var bibbase_data = {"data":"\"Loading..\"\n\n
\n\n \n\n \n\n \n \n\n \n\n \n \n\n \n\n \n
\n generated by\n \n \"bibbase.org\"\n\n \n
\n \n\n
\n\n \n\n\n
\n\n Excellent! Next you can\n create a new website with this list, or\n embed it in an existing web page by copying & pasting\n any of the following snippets.\n\n
\n JavaScript\n (easiest)\n
\n \n <script src=\"https://bibbase.org/show?bib=https%3A%2F%2Fbibbase.org%2Fzotero-group%2Fruasterry%2F2503580&jsonp=1&jsonp=1\"></script>\n \n
\n\n PHP\n
\n \n <?php\n $contents = file_get_contents(\"https://bibbase.org/show?bib=https%3A%2F%2Fbibbase.org%2Fzotero-group%2Fruasterry%2F2503580&jsonp=1\");\n print_r($contents);\n ?>\n \n
\n\n iFrame\n (not recommended)\n
\n \n <iframe src=\"https://bibbase.org/show?bib=https%3A%2F%2Fbibbase.org%2Fzotero-group%2Fruasterry%2F2503580&jsonp=1\"></iframe>\n \n
\n\n

\n For more details see the documention.\n

\n
\n
\n\n
\n\n This is a preview! To use this list on your own web site\n or create a new web site from it,\n create a free account. The file will be added\n and you will be able to edit it in the File Manager.\n We will show you instructions once you've created your account.\n
\n\n
\n\n

To the site owner:

\n\n

Action required! Mendeley is changing its\n API. In order to keep using Mendeley with BibBase past April\n 14th, you need to:\n

    \n
  1. renew the authorization for BibBase on Mendeley, and
    2. \n
  2. update the BibBase URL\n in your page the same way you did when you initially set up\n this page.\n
    3. \n
\n

\n\n

\n \n \n Fix it now\n

\n
\n\n
\n\n\n
\n \n \n
\n
\n  \n 2026\n \n \n (3)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Piecing Together Cross-Document Coreference Resolution Datasets: Systematic Dataset Analysis and Unification.\n \n \n \n \n\n\n \n Zhukova, A.; Ruas, T.; Wahle, J. P.; and Gipp, B.\n\n\n \n\n\n\n February 2026.\n arXiv:2603.00621 [cs]\n\n\n\n
\n\n\n\n \n \n \"PiecingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{zhukova_piecing_2026,\n\ttitle = {Piecing {Together} {Cross}-{Document} {Coreference} {Resolution} {Datasets}: {Systematic} {Dataset} {Analysis} and {Unification}},\n\tshorttitle = {Piecing {Together} {Cross}-{Document} {Coreference} {Resolution} {Datasets}},\n\turl = {http://arxiv.org/abs/2603.00621},\n\tdoi = {10.48550/arXiv.2603.00621},\n\tabstract = {Research in CDCR remains fragmented due to heterogeneous dataset formats, varying annotation standards, and the predominance of the CDCR definition as the event coreference resolution (ECR). To address these challenges, we introduce uCDCR, a unified dataset that consolidates diverse publicly available English CDCR corpora across various domains into a consistent format, which we analyze with standardized metrics and evaluation protocols. uCDCR incorporates both entity and event coreference, corrects known inconsistencies, and enriches datasets with missing attributes to facilitate reproducible research. We establish a cohesive framework for fair, interpretable, and cross-dataset analysis in CDCR and compare the datasets on their lexical properties, e.g., lexical composition of the annotated mentions, lexical diversity and ambiguity metrics, discuss the annotation rules and principles that lead to high lexical diversity, and examine how these metrics influence performance on the same-head-lemma baseline. Our dataset analysis shows that ECB+, the state-of-the-art benchmark for CDCR, has one of the lowest lexical diversities, and its CDCR complexity, measured by the same-head-lemma baseline, lies in the middle among all uCDCR datasets. Moreover, comparing document and mention distributions between ECB+ and uCDCR shows that using all uCDCR datasets for model training and evaluation will improve the generalizability of CDCR models. Finally, the almost identical performance on the same-head-lemma baseline, separately applied to events and entities, shows that resolving both types is a complex task and should not be steered toward ECR alone. The uCDCR dataset is available at https://huggingface.co/datasets/AnZhu/uCDCR, and the code for parsing, analyzing, and scoring the dataset is available at https://github.com/anastasia-zhukova/uCDCR.},\n\turldate = {2026-03-03},\n\tpublisher = {arXiv},\n\tauthor = {Zhukova, Anastasia and Ruas, Terry and Wahle, Jan Philip and Gipp, Bela},\n\tmonth = feb,\n\tyear = {2026},\n\tnote = {arXiv:2603.00621 [cs]},\n\tkeywords = {!tr, !tr\\_author, Computer Science - Computation and Language, nlp\\_cdcr},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Research in CDCR remains fragmented due to heterogeneous dataset formats, varying annotation standards, and the predominance of the CDCR definition as the event coreference resolution (ECR). To address these challenges, we introduce uCDCR, a unified dataset that consolidates diverse publicly available English CDCR corpora across various domains into a consistent format, which we analyze with standardized metrics and evaluation protocols. uCDCR incorporates both entity and event coreference, corrects known inconsistencies, and enriches datasets with missing attributes to facilitate reproducible research. We establish a cohesive framework for fair, interpretable, and cross-dataset analysis in CDCR and compare the datasets on their lexical properties, e.g., lexical composition of the annotated mentions, lexical diversity and ambiguity metrics, discuss the annotation rules and principles that lead to high lexical diversity, and examine how these metrics influence performance on the same-head-lemma baseline. Our dataset analysis shows that ECB+, the state-of-the-art benchmark for CDCR, has one of the lowest lexical diversities, and its CDCR complexity, measured by the same-head-lemma baseline, lies in the middle among all uCDCR datasets. Moreover, comparing document and mention distributions between ECB+ and uCDCR shows that using all uCDCR datasets for model training and evaluation will improve the generalizability of CDCR models. Finally, the almost identical performance on the same-head-lemma baseline, separately applied to events and entities, shows that resolving both types is a complex task and should not be steered toward ECR alone. The uCDCR dataset is available at https://huggingface.co/datasets/AnZhu/uCDCR, and the code for parsing, analyzing, and scoring the dataset is available at https://github.com/anastasia-zhukova/uCDCR.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n DimStance: Multilingual Datasets for Dimensional Stance Analysis.\n \n \n \n \n\n\n \n Becker, J.; Yu, L.; Muhammad, S. H.; Wahle, J. P.; Ruas, T.; Abdulmumin, I.; Lee, L.; Liu, W.; Lin, T.; Xu, Z.; Lin, Y.; Wang, J.; Mukhtar, M. I.; Gipp, B.; and Mohammed, S. M.\n\n\n \n\n\n\n January 2026.\n arXiv:2601.21483 [cs]\n\n\n\n
\n\n\n\n \n \n \"DimStance:Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{becker_dimstance_2026,\n\ttitle = {{DimStance}: {Multilingual} {Datasets} for {Dimensional} {Stance} {Analysis}},\n\tshorttitle = {{DimStance}},\n\turl = {http://arxiv.org/abs/2601.21483},\n\tdoi = {10.48550/arXiv.2601.21483},\n\tabstract = {Stance detection is an established task that classifies an author's attitude toward a specific target into categories such as Favor, Neutral, and Against. Beyond categorical stance labels, we leverage a long-established affective science framework to model stance along real-valued dimensions of valence (negative-positive) and arousal (calm-active). This dimensional approach captures nuanced affective states underlying stance expressions, enabling fine-grained stance analysis. To this end, we introduce DimStance, the first dimensional stance resource with valence-arousal (VA) annotations. This resource comprises 11,746 target aspects in 7,365 texts across five languages (English, German, Chinese, Nigerian Pidgin, and Swahili) and two domains (politics and environmental protection). To facilitate the evaluation of stance VA prediction, we formulate the dimensional stance regression task, analyze cross-lingual VA patterns, and benchmark pretrained and large language models under regression and prompting settings. Results show competitive performance of fine-tuned LLM regressors, persistent challenges in low-resource languages, and limitations of token-based generation. DimStance provides a foundation for multilingual, emotion-aware, stance analysis and benchmarking.},\n\turldate = {2026-02-04},\n\tpublisher = {arXiv},\n\tauthor = {Becker, Jonas and Yu, Liang-Chih and Muhammad, Shamsuddeen Hassan and Wahle, Jan Philip and Ruas, Terry and Abdulmumin, Idris and Lee, Lung-Hao and Liu, Wen-Ni and Lin, Tzu-Mi and Xu, Zhe-Yu and Lin, Ying-Lung and Wang, Jin and Mukhtar, Maryam Ibrahim and Gipp, Bela and Mohammed, Saif M.},\n\tmonth = jan,\n\tyear = {2026},\n\tnote = {arXiv:2601.21483 [cs]},\n\tkeywords = {!tr\\_author, Computer Science - Computation and Language, semeval},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Stance detection is an established task that classifies an author's attitude toward a specific target into categories such as Favor, Neutral, and Against. Beyond categorical stance labels, we leverage a long-established affective science framework to model stance along real-valued dimensions of valence (negative-positive) and arousal (calm-active). This dimensional approach captures nuanced affective states underlying stance expressions, enabling fine-grained stance analysis. To this end, we introduce DimStance, the first dimensional stance resource with valence-arousal (VA) annotations. This resource comprises 11,746 target aspects in 7,365 texts across five languages (English, German, Chinese, Nigerian Pidgin, and Swahili) and two domains (politics and environmental protection). To facilitate the evaluation of stance VA prediction, we formulate the dimensional stance regression task, analyze cross-lingual VA patterns, and benchmark pretrained and large language models under regression and prompting settings. Results show competitive performance of fine-tuned LLM regressors, persistent challenges in low-resource languages, and limitations of token-based generation. DimStance provides a foundation for multilingual, emotion-aware, stance analysis and benchmarking.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n DimABSA: Building Multilingual and Multidomain Datasets for Dimensional Aspect-Based Sentiment Analysis.\n \n \n \n \n\n\n \n Lee, L.; Yu, L.; Loukashevich, N.; Alimova, I.; Panchenko, A.; Lin, T.; Xu, Z.; Zhou, J.; Zheng, G.; Wang, J.; Awasthi, S.; Becker, J.; Wahle, J. P.; Ruas, T.; Muhammad, S. H.; and Mohammed, S. M.\n\n\n \n\n\n\n January 2026.\n arXiv:2601.23022 [cs]\n\n\n\n
\n\n\n\n \n \n \"DimABSA:Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{lee_dimabsa_2026,\n\ttitle = {{DimABSA}: {Building} {Multilingual} and {Multidomain} {Datasets} for {Dimensional} {Aspect}-{Based} {Sentiment} {Analysis}},\n\tshorttitle = {{DimABSA}},\n\turl = {http://arxiv.org/abs/2601.23022},\n\tdoi = {10.48550/arXiv.2601.23022},\n\tabstract = {Aspect-Based Sentiment Analysis (ABSA) focuses on extracting sentiment at a fine-grained aspect level and has been widely applied across real-world domains. However, existing ABSA research relies on coarse-grained categorical labels (e.g., positive, negative), which limits its ability to capture nuanced affective states. To address this limitation, we adopt a dimensional approach that represents sentiment with continuous valence-arousal (VA) scores, enabling fine-grained analysis at both the aspect and sentiment levels. To this end, we introduce DimABSA, the first multilingual, dimensional ABSA resource annotated with both traditional ABSA elements (aspect terms, aspect categories, and opinion terms) and newly introduced VA scores. This resource contains 76,958 aspect instances across 42,590 sentences, spanning six languages and four domains. We further introduce three subtasks that combine VA scores with different ABSA elements, providing a bridge from traditional ABSA to dimensional ABSA. Given that these subtasks involve both categorical and continuous outputs, we propose a new unified metric, continuous F1 (cF1), which incorporates VA prediction error into standard F1. We provide a comprehensive benchmark using both prompted and fine-tuned large language models across all subtasks. Our results show that DimABSA is a challenging benchmark and provides a foundation for advancing multilingual dimensional ABSA.},\n\turldate = {2026-02-04},\n\tpublisher = {arXiv},\n\tauthor = {Lee, Lung-Hao and Yu, Liang-Chih and Loukashevich, Natalia and Alimova, Ilseyar and Panchenko, Alexander and Lin, Tzu-Mi and Xu, Zhe-Yu and Zhou, Jian-Yu and Zheng, Guangmin and Wang, Jin and Awasthi, Sharanya and Becker, Jonas and Wahle, Jan Philip and Ruas, Terry and Muhammad, Shamsuddeen Hassan and Mohammed, Saif M.},\n\tmonth = jan,\n\tyear = {2026},\n\tnote = {arXiv:2601.23022 [cs]},\n\tkeywords = {!tr, Computer Science - Computation and Language, semeval},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Aspect-Based Sentiment Analysis (ABSA) focuses on extracting sentiment at a fine-grained aspect level and has been widely applied across real-world domains. However, existing ABSA research relies on coarse-grained categorical labels (e.g., positive, negative), which limits its ability to capture nuanced affective states. To address this limitation, we adopt a dimensional approach that represents sentiment with continuous valence-arousal (VA) scores, enabling fine-grained analysis at both the aspect and sentiment levels. To this end, we introduce DimABSA, the first multilingual, dimensional ABSA resource annotated with both traditional ABSA elements (aspect terms, aspect categories, and opinion terms) and newly introduced VA scores. This resource contains 76,958 aspect instances across 42,590 sentences, spanning six languages and four domains. We further introduce three subtasks that combine VA scores with different ABSA elements, providing a bridge from traditional ABSA to dimensional ABSA. Given that these subtasks involve both categorical and continuous outputs, we propose a new unified metric, continuous F1 (cF1), which incorporates VA prediction error into standard F1. We provide a comprehensive benchmark using both prompted and fine-tuned large language models across all subtasks. Our results show that DimABSA is a challenging benchmark and provides a foundation for advancing multilingual dimensional ABSA.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2025\n \n \n (18)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Big Tech-Funded AI Papers Have Higher Citation Impact, Greater Insularity, and Larger Recency Bias.\n \n \n \n \n\n\n \n Gnewuch, M. M.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n December 2025.\n arXiv:2512.05714 [cs]\n\n\n\n
\n\n\n\n \n \n \"BigPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{gnewuch_big_2025,\n\ttitle = {Big {Tech}-{Funded} {AI} {Papers} {Have} {Higher} {Citation} {Impact}, {Greater} {Insularity}, and {Larger} {Recency} {Bias}},\n\turl = {http://arxiv.org/abs/2512.05714},\n\tdoi = {10.48550/arXiv.2512.05714},\n\tabstract = {Over the past four decades, artificial intelligence (AI) research has flourished at the nexus of academia and industry. However, Big Tech companies have increasingly acquired the edge in computational resources, big data, and talent. So far, it has been largely unclear how many papers the industry funds, how their citation impact compares to non-funded papers, and what drives industry interest. This study fills that gap by quantifying the number of industry-funded papers at 10 top AI conferences (e.g., ICLR, CVPR, AAAI, ACL) and their citation influence. We analyze about 49.8K papers, about 1.8M citations from AI papers to other papers, and about 2.3M citations from other papers to AI papers from 1998-2022 in Scopus. Through seven research questions, we examine the volume and evolution of industry funding in AI research, the citation impact of funded papers, the diversity and temporal range of their citations, and the subfields in which industry predominantly acts. Our findings reveal that industry presence has grown markedly since 2015, from less than 2 percent to more than 11 percent in 2020. Between 2018 and 2022, 12 percent of industry-funded papers achieved high citation rates as measured by the h5-index, compared to 4 percent of non-industry-funded papers and 2 percent of non-funded papers. Top AI conferences engage more with industry-funded research than non-funded research, as measured by our newly proposed metric, the Citation Preference Ratio (CPR). We show that industry-funded research is increasingly insular, citing predominantly other industry-funded papers while referencing fewer non-funded papers. These findings reveal new trends in AI research funding, including a shift towards more industry-funded papers and their growing citation impact, greater insularity of industry-funded work than non-funded work, and a preference of industry-funded research to cite recent work.},\n\turldate = {2025-12-08},\n\tpublisher = {arXiv},\n\tauthor = {Gnewuch, Max Martin and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tmonth = dec,\n\tyear = {2025},\n\tnote = {arXiv:2512.05714 [cs]},\n\tkeywords = {!tr\\_author, Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Digital Libraries, scientometrics},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Over the past four decades, artificial intelligence (AI) research has flourished at the nexus of academia and industry. However, Big Tech companies have increasingly acquired the edge in computational resources, big data, and talent. So far, it has been largely unclear how many papers the industry funds, how their citation impact compares to non-funded papers, and what drives industry interest. This study fills that gap by quantifying the number of industry-funded papers at 10 top AI conferences (e.g., ICLR, CVPR, AAAI, ACL) and their citation influence. We analyze about 49.8K papers, about 1.8M citations from AI papers to other papers, and about 2.3M citations from other papers to AI papers from 1998-2022 in Scopus. Through seven research questions, we examine the volume and evolution of industry funding in AI research, the citation impact of funded papers, the diversity and temporal range of their citations, and the subfields in which industry predominantly acts. Our findings reveal that industry presence has grown markedly since 2015, from less than 2 percent to more than 11 percent in 2020. Between 2018 and 2022, 12 percent of industry-funded papers achieved high citation rates as measured by the h5-index, compared to 4 percent of non-industry-funded papers and 2 percent of non-funded papers. Top AI conferences engage more with industry-funded research than non-funded research, as measured by our newly proposed metric, the Citation Preference Ratio (CPR). We show that industry-funded research is increasingly insular, citing predominantly other industry-funded papers while referencing fewer non-funded papers. These findings reveal new trends in AI research funding, including a shift towards more industry-funded papers and their growing citation impact, greater insularity of industry-funded work than non-funded work, and a preference of industry-funded research to cite recent work.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n ALDEN: Reinforcement Learning for Active Navigation and Evidence Gathering in Long Documents.\n \n \n \n \n\n\n \n Yang, T.; Ruas, T.; Tian, Y.; Wahle, J. P.; Kurzawe, D.; and Gipp, B.\n\n\n \n\n\n\n October 2025.\n arXiv:2510.25668 [cs]\n\n\n\n
\n\n\n\n \n \n \"ALDEN:Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{yang_alden_2025,\n\ttitle = {{ALDEN}: {Reinforcement} {Learning} for {Active} {Navigation} and {Evidence} {Gathering} in {Long} {Documents}},\n\tshorttitle = {{ALDEN}},\n\turl = {http://arxiv.org/abs/2510.25668},\n\tdoi = {10.48550/arXiv.2510.25668},\n\tabstract = {Vision-language models (VLMs) excel at interpreting text-rich images but struggle with long, visually complex documents that demand analysis and integration of information spread across multiple pages. Existing approaches typically rely on fixed reasoning templates or rigid pipelines, which force VLMs into a passive role and hinder both efficiency and generalization. We present Active Long-DocumEnt Navigation (ALDEN), a multi-turn reinforcement learning framework that fine-tunes VLMs as interactive agents capable of actively navigating long, visually rich documents. ALDEN introduces a novel fetch action that directly accesses the page by index, complementing the classic search action and better exploiting document structure. For dense process supervision and efficient training, we propose a rule-based cross-level reward that provides both turn- and token-level signals. To address the empirically observed training instability caused by numerous visual tokens from long documents, we further propose a visual-semantic anchoring mechanism that applies a dual-path KL-divergence constraint to stabilize visual and textual representations separately during training. Trained on a corpus constructed from three open-source datasets, ALDEN achieves state-of-the-art performance on five long-document benchmarks. Overall, ALDEN marks a step beyond passive document reading toward agents that autonomously navigate and reason across long, visually rich documents, offering a robust path to more accurate and efficient long-document understanding.},\n\turldate = {2025-12-01},\n\tpublisher = {arXiv},\n\tauthor = {Yang, Tianyu and Ruas, Terry and Tian, Yijun and Wahle, Jan Philip and Kurzawe, Daniel and Gipp, Bela},\n\tmonth = oct,\n\tyear = {2025},\n\tnote = {arXiv:2510.25668 [cs]},\n\tkeywords = {!tr\\_author, Computer Science - Artificial Intelligence, Computer Science - Multimedia, ir, nlp\\_LLM, nlp\\_rl},\n}\n\n\n\n
\n
\n\n\n
\n Vision-language models (VLMs) excel at interpreting text-rich images but struggle with long, visually complex documents that demand analysis and integration of information spread across multiple pages. Existing approaches typically rely on fixed reasoning templates or rigid pipelines, which force VLMs into a passive role and hinder both efficiency and generalization. We present Active Long-DocumEnt Navigation (ALDEN), a multi-turn reinforcement learning framework that fine-tunes VLMs as interactive agents capable of actively navigating long, visually rich documents. ALDEN introduces a novel fetch action that directly accesses the page by index, complementing the classic search action and better exploiting document structure. For dense process supervision and efficient training, we propose a rule-based cross-level reward that provides both turn- and token-level signals. To address the empirically observed training instability caused by numerous visual tokens from long documents, we further propose a visual-semantic anchoring mechanism that applies a dual-path KL-divergence constraint to stabilize visual and textual representations separately during training. Trained on a corpus constructed from three open-source datasets, ALDEN achieves state-of-the-art performance on five long-document benchmarks. Overall, ALDEN marks a step beyond passive document reading toward agents that autonomously navigate and reason across long, visually rich documents, offering a robust path to more accurate and efficient long-document understanding.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Overview of the Plagiarism Detection Task at PAN 2025.\n \n \n \n \n\n\n \n Greiner-Petter, A.; Fröbe, M.; Wahle, J. P.; Ruas, T.; Gipp, B.; Aizawa, A.; and Potthast, M.\n\n\n \n\n\n\n October 2025.\n arXiv:2510.06805 [cs]\n\n\n\n
\n\n\n\n \n \n \"OverviewPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{greiner-petter_overview_2025,\n\ttitle = {Overview of the {Plagiarism} {Detection} {Task} at {PAN} 2025},\n\turl = {http://arxiv.org/abs/2510.06805},\n\tdoi = {10.48550/arXiv.2510.06805},\n\tabstract = {The generative plagiarism detection task at PAN 2025 aims at identifying automatically generated textual plagiarism in scientific articles and aligning them with their respective sources. We created a novel large-scale dataset of automatically generated plagiarism using three large language models: Llama, DeepSeek-R1, and Mistral. In this task overview paper, we outline the creation of this dataset, summarize and compare the results of all participants and four baselines, and evaluate the results on the last plagiarism detection task from PAN 2015 in order to interpret the robustness of the proposed approaches. We found that the current iteration does not invite a large variety of approaches as naive semantic similarity approaches based on embedding vectors provide promising results of up to 0.8 recall and 0.5 precision. In contrast, most of these approaches underperform significantly on the 2015 dataset, indicating a lack in generalizability.},\n\turldate = {2025-12-01},\n\tpublisher = {arXiv},\n\tauthor = {Greiner-Petter, André and Fröbe, Maik and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela and Aizawa, Akiko and Potthast, Martin},\n\tmonth = oct,\n\tyear = {2025},\n\tnote = {arXiv:2510.06805 [cs]},\n\tkeywords = {!tr\\_author, Computer Science - Computation and Language, Computer Science - Information Retrieval, nlp\\_plagiarism, workshop},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n The generative plagiarism detection task at PAN 2025 aims at identifying automatically generated textual plagiarism in scientific articles and aligning them with their respective sources. We created a novel large-scale dataset of automatically generated plagiarism using three large language models: Llama, DeepSeek-R1, and Mistral. In this task overview paper, we outline the creation of this dataset, summarize and compare the results of all participants and four baselines, and evaluate the results on the last plagiarism detection task from PAN 2015 in order to interpret the robustness of the proposed approaches. We found that the current iteration does not invite a large variety of approaches as naive semantic similarity approaches based on embedding vectors provide promising results of up to 0.8 recall and 0.5 precision. In contrast, most of these approaches underperform significantly on the 2015 dataset, indicating a lack in generalizability.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n TrojanStego: Your Language Model Can Secretly Be A Steganographic Privacy Leaking Agent.\n \n \n \n \n\n\n \n Meier, D.; Wahle, J. P.; Röttger, P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 27232–27249, Suzhou, China, 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"TrojanStego:Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{meier_trojanstego_2025,\n\taddress = {Suzhou, China},\n\ttitle = {{TrojanStego}: {Your} {Language} {Model} {Can} {Secretly} {Be} {A} {Steganographic} {Privacy} {Leaking} {Agent}},\n\tshorttitle = {{TrojanStego}},\n\turl = {https://aclanthology.org/2025.emnlp-main.1386},\n\tdoi = {10.18653/v1/2025.emnlp-main.1386},\n\tlanguage = {en},\n\turldate = {2025-11-11},\n\tbooktitle = {Proceedings of the 2025 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Meier, Dominik and Wahle, Jan Philip and Röttger, Paul and Ruas, Terry and Gipp, Bela},\n\tyear = {2025},\n\tpages = {27232--27249},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Re-FRAME the Meeting Summarization SCOPE: Fact-Based Summarization and Personalization via Questions.\n \n \n \n \n\n\n \n Kirstein, F.; Kumar, S.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 20087–20137, Suzhou, China, 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"Re-FRAMEPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{kirstein_re-frame_2025,\n\taddress = {Suzhou, China},\n\ttitle = {Re-{FRAME} the {Meeting} {Summarization} {SCOPE}: {Fact}-{Based} {Summarization} and {Personalization} via {Questions}},\n\tshorttitle = {Re-{FRAME} the {Meeting} {Summarization} {SCOPE}},\n\turl = {https://aclanthology.org/2025.findings-emnlp.1094},\n\tdoi = {10.18653/v1/2025.findings-emnlp.1094},\n\tlanguage = {en},\n\turldate = {2025-11-11},\n\tbooktitle = {Findings of the {Association} for {Computational} {Linguistics}: {EMNLP} 2025},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kirstein, Frederic and Kumar, Sonu and Ruas, Terry and Gipp, Bela},\n\tyear = {2025},\n\tpages = {20087--20137},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n SPaRC: A Spatial Pathfinding Reasoning Challenge.\n \n \n \n \n\n\n \n Kaesberg, L. B.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 10370–10401, Suzhou, China, 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"SPaRC:Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{kaesberg_sparc_2025,\n\taddress = {Suzhou, China},\n\ttitle = {{SPaRC}: {A} {Spatial} {Pathfinding} {Reasoning} {Challenge}},\n\tshorttitle = {{SPaRC}},\n\turl = {https://aclanthology.org/2025.emnlp-main.526},\n\tdoi = {10.18653/v1/2025.emnlp-main.526},\n\tlanguage = {en},\n\turldate = {2025-11-11},\n\tbooktitle = {Proceedings of the 2025 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kaesberg, Lars Benedikt and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tyear = {2025},\n\tpages = {10370--10401},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n MALLM: Multi-Agent Large Language Models Framework.\n \n \n \n \n\n\n \n Becker, J.; Kaesberg, L. B.; Bauer, N.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 418–439, Suzhou, China, 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"MALLM:Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{becker_mallm_2025,\n\taddress = {Suzhou, China},\n\ttitle = {{MALLM}: {Multi}-{Agent} {Large} {Language} {Models} {Framework}},\n\tshorttitle = {{MALLM}},\n\turl = {https://aclanthology.org/2025.emnlp-demos.29},\n\tdoi = {10.18653/v1/2025.emnlp-demos.29},\n\tlanguage = {en},\n\turldate = {2025-11-11},\n\tbooktitle = {Proceedings of the 2025 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}: {System} {Demonstrations}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Becker, Jonas and Kaesberg, Lars Benedikt and Bauer, Niklas and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tyear = {2025},\n\tpages = {418--439},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n You need to MIMIC to get FAME: Solving Meeting Transcript Scarcity with Multi-Agent Conversations.\n \n \n \n \n\n\n \n Kirstein, F.; Khan, M.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Findings of the Association for Computational Linguistics: ACL 2025, pages 11482–11525, Vienna, Austria, 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"YouPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{kirstein_you_2025,\n\taddress = {Vienna, Austria},\n\ttitle = {You need to {MIMIC} to get {FAME}: {Solving} {Meeting} {Transcript} {Scarcity} with {Multi}-{Agent} {Conversations}},\n\tshorttitle = {You need to {MIMIC} to get {FAME}},\n\turl = {https://aclanthology.org/2025.findings-acl.599},\n\tdoi = {10.18653/v1/2025.findings-acl.599},\n\tlanguage = {en},\n\turldate = {2025-09-01},\n\tbooktitle = {Findings of the {Association} for {Computational} {Linguistics}: {ACL} 2025},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kirstein, Frederic and Khan, Muneeb and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tyear = {2025},\n\tpages = {11482--11525},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Voting or Consensus? Decision-Making in Multi-Agent Debate.\n \n \n \n \n\n\n \n Kaesberg, L. B.; Becker, J.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Findings of the Association for Computational Linguistics: ACL 2025, pages 11640–11671, Vienna, Austria, 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"VotingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{kaesberg_voting_2025,\n\taddress = {Vienna, Austria},\n\ttitle = {Voting or {Consensus}? {Decision}-{Making} in {Multi}-{Agent} {Debate}},\n\tshorttitle = {Voting or {Consensus}?},\n\turl = {https://aclanthology.org/2025.findings-acl.606},\n\tdoi = {10.18653/v1/2025.findings-acl.606},\n\tlanguage = {en},\n\turldate = {2025-09-01},\n\tbooktitle = {Findings of the {Association} for {Computational} {Linguistics}: {ACL} 2025},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kaesberg, Lars Benedikt and Becker, Jonas and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tyear = {2025},\n\tpages = {11640--11671},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n BRIGHTER: BRIdging the Gap in Human-Annotated Textual Emotion Recognition Datasets for 28 Languages.\n \n \n \n \n\n\n \n Muhammad, S. H.; Ousidhoum, N.; Abdulmumin, I.; Wahle, J. P.; Ruas, T.; Beloucif, M.; De Kock, C.; Surange, N.; Teodorescu, D.; Ahmad, I. S.; Adelani, D. I.; Aji, A. F.; Ali, F. D. M. A.; Alimova, I.; Araujo, V.; Babakov, N.; Baes, N.; Bucur, A.; Bukula, A.; Cao, G.; Tufiño, R.; Chevi, R.; Chukwuneke, C. I.; Ciobotaru, A.; Dementieva, D.; Gadanya, M. S.; Geislinger, R.; Gipp, B.; Hourrane, O.; Ignat, O.; Lawan, F. I.; Mabuya, R.; Mahendra, R.; Marivate, V.; Panchenko, A.; Piper, A.; Ferreira, C. H. P.; Protasov, V.; Rutunda, S.; Shrivastava, M.; Udrea, A. C.; Wanzare, L. D. A.; Wu, S.; Wunderlich, F. V.; Zhafran, H. M.; Zhang, T.; Zhou, Y.; and Mohammad, S. M.\n\n\n \n\n\n\n In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8895–8916, Vienna, Austria, 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"BRIGHTER:Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{muhammad_brighter_2025,\n\taddress = {Vienna, Austria},\n\ttitle = {{BRIGHTER}: {BRIdging} the {Gap} in {Human}-{Annotated} {Textual} {Emotion} {Recognition} {Datasets} for 28 {Languages}},\n\tshorttitle = {{BRIGHTER}},\n\turl = {https://aclanthology.org/2025.acl-long.436},\n\tdoi = {10.18653/v1/2025.acl-long.436},\n\tlanguage = {en},\n\turldate = {2025-09-01},\n\tbooktitle = {Proceedings of the 63rd {Annual} {Meeting} of the {Association} for {Computational} {Linguistics} ({Volume} 1: {Long} {Papers})},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Muhammad, Shamsuddeen Hassan and Ousidhoum, Nedjma and Abdulmumin, Idris and Wahle, Jan Philip and Ruas, Terry and Beloucif, Meriem and De Kock, Christine and Surange, Nirmal and Teodorescu, Daniela and Ahmad, Ibrahim Said and Adelani, David Ifeoluwa and Aji, Alham Fikri and Ali, Felermino D. M. A. and Alimova, Ilseyar and Araujo, Vladimir and Babakov, Nikolay and Baes, Naomi and Bucur, Ana-Maria and Bukula, Andiswa and Cao, Guanqun and Tufiño, Rodrigo and Chevi, Rendi and Chukwuneke, Chiamaka Ijeoma and Ciobotaru, Alexandra and Dementieva, Daryna and Gadanya, Murja Sani and Geislinger, Robert and Gipp, Bela and Hourrane, Oumaima and Ignat, Oana and Lawan, Falalu Ibrahim and Mabuya, Rooweither and Mahendra, Rahmad and Marivate, Vukosi and Panchenko, Alexander and Piper, Andrew and Ferreira, Charles Henrique Porto and Protasov, Vitaly and Rutunda, Samuel and Shrivastava, Manish and Udrea, Aura Cristina and Wanzare, Lilian Diana Awuor and Wu, Sophie and Wunderlich, Florian Valentin and Zhafran, Hanif Muhammad and Zhang, Tianhui and Zhou, Yi and Mohammad, Saif M.},\n\tyear = {2025},\n\tpages = {8895--8916},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Efficient Domain-adaptive Continual Pretraining for the Process Industry in the German Language.\n \n \n \n \n\n\n \n Zhukova, A.; Matt, C. E.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n April 2025.\n arXiv:2504.19856 [cs]\n\n\n\n
\n\n\n\n \n \n \"EfficientPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{zhukova_efficient_2025,\n\ttitle = {Efficient {Domain}-adaptive {Continual} {Pretraining} for the {Process} {Industry} in the {German} {Language}},\n\turl = {http://arxiv.org/abs/2504.19856},\n\tdoi = {10.48550/arXiv.2504.19856},\n\tabstract = {Domain-adaptive continual pretraining (DAPT) is a state-of-the-art technique that further trains a language model (LM) on its pretraining task, e.g., language masking. Although popular, it requires a significant corpus of domain-related data, which is difficult to obtain for specific domains in languages other than English, such as the process industry in the German language. This paper introduces an efficient approach called ICL-augmented pretraining or ICL-APT that leverages in-context learning (ICL) and k-nearest neighbors (kNN) to augment target data with domain-related and in-domain texts, significantly reducing GPU time while maintaining strong model performance. Our results show that this approach performs better than traditional DAPT by 3.5 points of the average IR metrics (e.g., mAP, MRR, and nDCG) and requires almost 4 times less computing time, providing a cost-effective solution for industries with limited computational capacity. The findings highlight the broader applicability of this framework to other low-resource industries, making NLP-based solutions more accessible and feasible in production environments.},\n\turldate = {2025-06-25},\n\tpublisher = {arXiv},\n\tauthor = {Zhukova, Anastasia and Matt, Christian E. and Ruas, Terry and Gipp, Bela},\n\tmonth = apr,\n\tyear = {2025},\n\tnote = {arXiv:2504.19856 [cs]},\n\tkeywords = {!tr\\_author, Computer Science - Computation and Language, nlp\\_domain\\_adaptation, nlp\\_plant\\_assist},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Domain-adaptive continual pretraining (DAPT) is a state-of-the-art technique that further trains a language model (LM) on its pretraining task, e.g., language masking. Although popular, it requires a significant corpus of domain-related data, which is difficult to obtain for specific domains in languages other than English, such as the process industry in the German language. This paper introduces an efficient approach called ICL-augmented pretraining or ICL-APT that leverages in-context learning (ICL) and k-nearest neighbors (kNN) to augment target data with domain-related and in-domain texts, significantly reducing GPU time while maintaining strong model performance. Our results show that this approach performs better than traditional DAPT by 3.5 points of the average IR metrics (e.g., mAP, MRR, and nDCG) and requires almost 4 times less computing time, providing a cost-effective solution for industries with limited computational capacity. The findings highlight the broader applicability of this framework to other low-resource industries, making NLP-based solutions more accessible and feasible in production environments.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n SemEval-2025 Task 11: Bridging the Gap in Text-Based Emotion Detection.\n \n \n \n \n\n\n \n Muhammad, S. H.; Ousidhoum, N.; Abdulmumin, I.; Yimam, S. M.; Wahle, J. P.; Ruas, T.; Beloucif, M.; Kock, C. D.; Belay, T. D.; Ahmad, I. S.; Surange, N.; Teodorescu, D.; Adelani, D. I.; Aji, A. F.; Ali, F.; Araujo, V.; Ayele, A. A.; Ignat, O.; Panchenko, A.; Zhou, Y.; and Mohammad, S. M.\n\n\n \n\n\n\n March 2025.\n arXiv:2503.07269 [cs]\n\n\n\n
\n\n\n\n \n \n \"SemEval-2025Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{muhammad_semeval-2025_2025,\n\ttitle = {{SemEval}-2025 {Task} 11: {Bridging} the {Gap} in {Text}-{Based} {Emotion} {Detection}},\n\tshorttitle = {{SemEval}-2025 {Task} 11},\n\turl = {http://arxiv.org/abs/2503.07269},\n\tdoi = {10.48550/arXiv.2503.07269},\n\tabstract = {We present our shared task on text-based emotion detection, covering more than 30 languages from seven distinct language families. These languages are predominantly low-resource and spoken across various continents. The data instances are multi-labeled into six emotional classes, with additional datasets in 11 languages annotated for emotion intensity. Participants were asked to predict labels in three tracks: (a) emotion labels in monolingual settings, (b) emotion intensity scores, and (c) emotion labels in cross-lingual settings. The task attracted over 700 participants. We received final submissions from more than 200 teams and 93 system description papers. We report baseline results, as well as findings on the best-performing systems, the most common approaches, and the most effective methods across various tracks and languages. The datasets for this task are publicly available.},\n\turldate = {2025-03-11},\n\tpublisher = {arXiv},\n\tauthor = {Muhammad, Shamsuddeen Hassan and Ousidhoum, Nedjma and Abdulmumin, Idris and Yimam, Seid Muhie and Wahle, Jan Philip and Ruas, Terry and Beloucif, Meriem and Kock, Christine De and Belay, Tadesse Destaw and Ahmad, Ibrahim Said and Surange, Nirmal and Teodorescu, Daniela and Adelani, David Ifeoluwa and Aji, Alham Fikri and Ali, Felermino and Araujo, Vladimir and Ayele, Abinew Ali and Ignat, Oana and Panchenko, Alexander and Zhou, Yi and Mohammad, Saif M.},\n\tmonth = mar,\n\tyear = {2025},\n\tnote = {arXiv:2503.07269 [cs]},\n\tkeywords = {!tr\\_author, Computer Science - Computation and Language, semeval},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n We present our shared task on text-based emotion detection, covering more than 30 languages from seven distinct language families. These languages are predominantly low-resource and spoken across various continents. The data instances are multi-labeled into six emotional classes, with additional datasets in 11 languages annotated for emotion intensity. Participants were asked to predict labels in three tracks: (a) emotion labels in monolingual settings, (b) emotion intensity scores, and (c) emotion labels in cross-lingual settings. The task attracted over 700 participants. We received final submissions from more than 200 teams and 93 system description papers. We report baseline results, as well as findings on the best-performing systems, the most common approaches, and the most effective methods across various tracks and languages. The datasets for this task are publicly available.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Is my Meeting Summary Good? Estimating Quality with a Multi-LLM Evaluator.\n \n \n \n \n\n\n \n Kirstein, F. T.; Lima Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Rambow, O.; Wanner, L.; Apidianaki, M.; Al-Khalifa, H.; Eugenio, B. D.; Schockaert, S.; Darwish, K.; and Agarwal, A., editor(s), Proceedings of the 31st International Conference on Computational Linguistics: Industry Track, pages 561–574, Abu Dhabi, UAE, January 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"IsPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 2 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{kirstein-etal-2025-meeting,\n\taddress = {Abu Dhabi, UAE},\n\ttitle = {Is my {Meeting} {Summary} {Good}? {Estimating} {Quality} with a {Multi}-{LLM} {Evaluator}},\n\turl = {https://aclanthology.org/2025.coling-industry.48/},\n\tabstract = {The quality of meeting summaries generated by natural language generation (NLG) systems is hard to measure automatically. Established metrics such as ROUGE and BERTScore have a relatively low correlation with human judgments and fail to capture nuanced errors. Recent studies suggest using large language models (LLMs), which have the benefit of better context understanding and adaption of error definitions without training on a large number of human preference judgments. However, current LLM-based evaluators risk masking errors and can only serve as a weak proxy, leaving human evaluation the gold standard despite being costly and hard to compare across studies. In this work, we present MESA, an LLM-based framework employing a three-step assessment of individual error types, multi-agent discussion for decision refinement, and feedback-based self-training to refine error definition understanding and alignment with human judgment. We show that MESA`s components enable thorough error detection, consistent rating, and adaptability to custom error guidelines. Using GPT-4o as its backbone, MESA achieves mid to high Point-Biserial correlation with human judgment in error detection and mid Spearman and Kendall correlation in reflecting error impact on summary quality, on average 0.25 higher than previous methods. The framework`s flexibility in adapting to custom error guidelines makes it suitable for various tasks with limited human-labeled data.},\n\tbooktitle = {Proceedings of the 31st {International} {Conference} on {Computational} {Linguistics}: {Industry} {Track}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kirstein, Frederic Thomas and Lima Ruas, Terry and Gipp, Bela},\n\teditor = {Rambow, Owen and Wanner, Leo and Apidianaki, Marianna and Al-Khalifa, Hend and Eugenio, Barbara Di and Schockaert, Steven and Darwish, Kareem and Agarwal, Apoorv},\n\tmonth = jan,\n\tyear = {2025},\n\tpages = {561--574},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n The quality of meeting summaries generated by natural language generation (NLG) systems is hard to measure automatically. Established metrics such as ROUGE and BERTScore have a relatively low correlation with human judgments and fail to capture nuanced errors. Recent studies suggest using large language models (LLMs), which have the benefit of better context understanding and adaption of error definitions without training on a large number of human preference judgments. However, current LLM-based evaluators risk masking errors and can only serve as a weak proxy, leaving human evaluation the gold standard despite being costly and hard to compare across studies. In this work, we present MESA, an LLM-based framework employing a three-step assessment of individual error types, multi-agent discussion for decision refinement, and feedback-based self-training to refine error definition understanding and alignment with human judgment. We show that MESA`s components enable thorough error detection, consistent rating, and adaptability to custom error guidelines. Using GPT-4o as its backbone, MESA achieves mid to high Point-Biserial correlation with human judgment in error detection and mid Spearman and Kendall correlation in reflecting error impact on summary quality, on average 0.25 higher than previous methods. The framework`s flexibility in adapting to custom error guidelines makes it suitable for various tasks with limited human-labeled data.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Towards Human Understanding of Paraphrase Types in Large Language Models.\n \n \n \n \n\n\n \n Meier, D.; Wahle, J. P.; Lima Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Rambow, O.; Wanner, L.; Apidianaki, M.; Al-Khalifa, H.; Eugenio, B. D.; and Schockaert, S., editor(s), Proceedings of the 31st International Conference on Computational Linguistics, pages 6298–6316, Abu Dhabi, UAE, January 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"TowardsPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{meier-etal-2025-towards,\n\taddress = {Abu Dhabi, UAE},\n\ttitle = {Towards {Human} {Understanding} of {Paraphrase} {Types} in {Large} {Language} {Models}},\n\turl = {https://aclanthology.org/2025.coling-main.421/},\n\tabstract = {Paraphrases represent a human`s intuitive ability to understand expressions presented in various different ways. Current paraphrase evaluations of language models primarily use binary approaches, offering limited interpretability of specific text changes. Atomic paraphrase types (APT) decompose paraphrases into different linguistic changes and offer a granular view of the flexibility in linguistic expression (e.g., a shift in syntax or vocabulary used). In this study, we assess the human preferences towards ChatGPT in generating English paraphrases with ten APTs and five prompting techniques. We introduce APTY (Atomic Paraphrase TYpes), a dataset of 800 sentence-level and word-level annotations by 15 annotators. The dataset also provides a human preference ranking of paraphrases with different types that can be used to fine-tune models with RLHF and DPO methods. Our results reveal that ChatGPT and a DPO-trained LLama 7B model can generate simple APTs, such as additions and deletions, but struggle with complex structures (e.g., subordination changes). This study contributes to understanding which aspects of paraphrasing language models have already succeeded at understanding and what remains elusive. In addition, we show how our curated datasets can be used to develop language models with specific linguistic capabilities.},\n\tbooktitle = {Proceedings of the 31st {International} {Conference} on {Computational} {Linguistics}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Meier, Dominik and Wahle, Jan Philip and Lima Ruas, Terry and Gipp, Bela},\n\teditor = {Rambow, Owen and Wanner, Leo and Apidianaki, Marianna and Al-Khalifa, Hend and Eugenio, Barbara Di and Schockaert, Steven},\n\tmonth = jan,\n\tyear = {2025},\n\tpages = {6298--6316},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Paraphrases represent a human`s intuitive ability to understand expressions presented in various different ways. Current paraphrase evaluations of language models primarily use binary approaches, offering limited interpretability of specific text changes. Atomic paraphrase types (APT) decompose paraphrases into different linguistic changes and offer a granular view of the flexibility in linguistic expression (e.g., a shift in syntax or vocabulary used). In this study, we assess the human preferences towards ChatGPT in generating English paraphrases with ten APTs and five prompting techniques. We introduce APTY (Atomic Paraphrase TYpes), a dataset of 800 sentence-level and word-level annotations by 15 annotators. The dataset also provides a human preference ranking of paraphrases with different types that can be used to fine-tune models with RLHF and DPO methods. Our results reveal that ChatGPT and a DPO-trained LLama 7B model can generate simple APTs, such as additions and deletions, but struggle with complex structures (e.g., subordination changes). This study contributes to understanding which aspects of paraphrasing language models have already succeeded at understanding and what remains elusive. In addition, we show how our curated datasets can be used to develop language models with specific linguistic capabilities.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n What`s Wrong? Refining Meeting Summaries with LLM Feedback.\n \n \n \n \n\n\n \n Kirstein, F. T.; Lima Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Rambow, O.; Wanner, L.; Apidianaki, M.; Al-Khalifa, H.; Eugenio, B. D.; and Schockaert, S., editor(s), Proceedings of the 31st International Conference on Computational Linguistics, pages 2100–2120, Abu Dhabi, UAE, January 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"What`sPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{kirstein-etal-2025-whats,\n\taddress = {Abu Dhabi, UAE},\n\ttitle = {What`s {Wrong}? {Refining} {Meeting} {Summaries} with {LLM} {Feedback}},\n\turl = {https://aclanthology.org/2025.coling-main.143/},\n\tabstract = {Meeting summarization has become a critical task since digital encounters have become a common practice. Large language models (LLMs) show great potential in summarization, offering enhanced coherence and context understanding compared to traditional methods. However, they still struggle to maintain relevance and avoid hallucination. We introduce a multi-LLM correction approach for meeting summarization using a two-phase process that mimics the human review process: mistake identification and summary refinement. We release QMSum Mistake, a dataset of 200 automatically generated meeting summaries annotated by humans on nine error types, including structural, omission, and irrelevance errors. Our experiments show that these errors can be identified with high accuracy by an LLM. We transform identified mistakes into actionable feedback to improve the quality of a given summary measured by relevance, informativeness, conciseness, and coherence. This post-hoc refinement effectively improves summary quality by leveraging multiple LLMs to validate output quality. Our multi-LLM approach for meeting summarization shows potential for similar complex text generation tasks requiring robustness, action planning, and discussion towards a goal.},\n\tbooktitle = {Proceedings of the 31st {International} {Conference} on {Computational} {Linguistics}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kirstein, Frederic Thomas and Lima Ruas, Terry and Gipp, Bela},\n\teditor = {Rambow, Owen and Wanner, Leo and Apidianaki, Marianna and Al-Khalifa, Hend and Eugenio, Barbara Di and Schockaert, Steven},\n\tmonth = jan,\n\tyear = {2025},\n\tpages = {2100--2120},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Meeting summarization has become a critical task since digital encounters have become a common practice. Large language models (LLMs) show great potential in summarization, offering enhanced coherence and context understanding compared to traditional methods. However, they still struggle to maintain relevance and avoid hallucination. We introduce a multi-LLM correction approach for meeting summarization using a two-phase process that mimics the human review process: mistake identification and summary refinement. We release QMSum Mistake, a dataset of 200 automatically generated meeting summaries annotated by humans on nine error types, including structural, omission, and irrelevance errors. Our experiments show that these errors can be identified with high accuracy by an LLM. We transform identified mistakes into actionable feedback to improve the quality of a given summary measured by relevance, informativeness, conciseness, and coherence. This post-hoc refinement effectively improves summary quality by leveraging multiple LLMs to validate output quality. Our multi-LLM approach for meeting summarization shows potential for similar complex text generation tasks requiring robustness, action planning, and discussion towards a goal.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Citation Amnesia: On The Recency Bias of NLP and Other Academic Fields.\n \n \n \n \n\n\n \n Wahle, J. P.; Lima Ruas, T.; Abdalla, M.; Gipp, B.; and Mohammad, S. M.\n\n\n \n\n\n\n In Rambow, O.; Wanner, L.; Apidianaki, M.; Al-Khalifa, H.; Eugenio, B. D.; and Schockaert, S., editor(s), Proceedings of the 31st International Conference on Computational Linguistics, pages 1027–1044, Abu Dhabi, UAE, January 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"CitationPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{wahle-etal-2025-citation,\n\taddress = {Abu Dhabi, UAE},\n\ttitle = {Citation {Amnesia}: {On} {The} {Recency} {Bias} of {NLP} and {Other} {Academic} {Fields}},\n\turl = {https://aclanthology.org/2025.coling-main.69/},\n\tabstract = {This study examines the tendency to cite older work across 20 fields of study over 43 years (1980–2023). We put NLP`s propensity to cite older work in the context of these 20 other fields to analyze whether NLP shows similar temporal citation patterns to them over time or whether differences can be observed. Our analysis, based on a dataset of {\\textasciitilde}240 million papers, reveals a broader scientific trend: many fields have markedly declined in citing older works (e.g., psychology, computer science). The trend is strongest in NLP and ML research (-12.8\\% and -5.5\\% in citation age from previous peaks). Our results suggest that citing more recent works is not directly driven by the growth in publication rates (-3.4\\% across fields; -5.2\\% in humanities; -5.5\\% in formal sciences) — even when controlling for an increase in the volume of papers. Our findings raise questions about the scientific community`s engagement with past literature, particularly for NLP, and the potential consequences of neglecting older but relevant research. The data and a demo showcasing our results are publicly available.},\n\tbooktitle = {Proceedings of the 31st {International} {Conference} on {Computational} {Linguistics}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Wahle, Jan Philip and Lima Ruas, Terry and Abdalla, Mohamed and Gipp, Bela and Mohammad, Saif M.},\n\teditor = {Rambow, Owen and Wanner, Leo and Apidianaki, Marianna and Al-Khalifa, Hend and Eugenio, Barbara Di and Schockaert, Steven},\n\tmonth = jan,\n\tyear = {2025},\n\tpages = {1027--1044},\n}\n\n\n\n
\n
\n\n\n
\n This study examines the tendency to cite older work across 20 fields of study over 43 years (1980–2023). We put NLP`s propensity to cite older work in the context of these 20 other fields to analyze whether NLP shows similar temporal citation patterns to them over time or whether differences can be observed. Our analysis, based on a dataset of ~240 million papers, reveals a broader scientific trend: many fields have markedly declined in citing older works (e.g., psychology, computer science). The trend is strongest in NLP and ML research (-12.8% and -5.5% in citation age from previous peaks). Our results suggest that citing more recent works is not directly driven by the growth in publication rates (-3.4% across fields; -5.2% in humanities; -5.5% in formal sciences) — even when controlling for an increase in the volume of papers. Our findings raise questions about the scientific community`s engagement with past literature, particularly for NLP, and the potential consequences of neglecting older but relevant research. The data and a demo showcasing our results are publicly available.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Stay Focused: Problem Drift in Multi-Agent Debate.\n \n \n \n \n\n\n \n Becker, J.; Kaesberg, L. B.; Stephan, A.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n February 2025.\n arXiv:2502.19559 [cs]\n\n\n\n
\n\n\n\n \n \n \"StayPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 2 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{becker_stay_2025,\n\ttitle = {Stay {Focused}: {Problem} {Drift} in {Multi}-{Agent} {Debate}},\n\tshorttitle = {Stay {Focused}},\n\turl = {http://arxiv.org/abs/2502.19559},\n\tdoi = {10.48550/arXiv.2502.19559},\n\tabstract = {Multi-agent debate - multiple instances of large language models discussing problems in turn-based interaction - has shown promise for solving knowledge and reasoning tasks. However, these methods show limitations, particularly when scaling them to longer reasoning chains. In this study, we unveil a new issue of multi-agent debate: discussions drift away from the initial problem over multiple turns. We define this phenomenon as problem drift and quantify its presence across ten tasks (i.e., three generative, three knowledge, three reasoning, and one instruction-following task). To identify the reasons for this issue, we perform a human study with eight experts on discussions suffering from problem drift, who find the most common issues are a lack of progress (35\\% of cases), low-quality feedback (26\\% of cases), and a lack of clarity (25\\% of cases). To systematically address the issue of problem drift, we propose DRIFTJudge, a method based on LLM-as-a-judge, to detect problem drift at test-time. We further propose DRIFTPolicy, a method to mitigate 31\\% of problem drift cases. Our study can be seen as a first step to understanding a key limitation of multi-agent debate, highlighting pathways for improving their effectiveness in the future.},\n\turldate = {2025-02-28},\n\tpublisher = {arXiv},\n\tauthor = {Becker, Jonas and Kaesberg, Lars Benedikt and Stephan, Andreas and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tmonth = feb,\n\tyear = {2025},\n\tnote = {arXiv:2502.19559 [cs]},\n\tkeywords = {!tr, Computer Science - Computation and Language, nlp\\_llm, nlp\\_multiagent},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Multi-agent debate - multiple instances of large language models discussing problems in turn-based interaction - has shown promise for solving knowledge and reasoning tasks. However, these methods show limitations, particularly when scaling them to longer reasoning chains. In this study, we unveil a new issue of multi-agent debate: discussions drift away from the initial problem over multiple turns. We define this phenomenon as problem drift and quantify its presence across ten tasks (i.e., three generative, three knowledge, three reasoning, and one instruction-following task). To identify the reasons for this issue, we perform a human study with eight experts on discussions suffering from problem drift, who find the most common issues are a lack of progress (35% of cases), low-quality feedback (26% of cases), and a lack of clarity (25% of cases). To systematically address the issue of problem drift, we propose DRIFTJudge, a method based on LLM-as-a-judge, to detect problem drift at test-time. We further propose DRIFTPolicy, a method to mitigate 31% of problem drift cases. Our study can be seen as a first step to understanding a key limitation of multi-agent debate, highlighting pathways for improving their effectiveness in the future.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n CADS: A Systematic Literature Review on the Challenges of Abstractive Dialogue Summarization.\n \n \n \n \n\n\n \n Kirstein, F.; Wahle, J. P.; Gipp, B.; and Ruas, T.\n\n\n \n\n\n\n Journal of Artificial Intelligence Research, 82: 313–365. January 2025.\n \n\n\n\n
\n\n\n\n \n \n \"CADS:Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@article{kirstein_cads_2025,\n\ttitle = {{CADS}: {A} {Systematic} {Literature} {Review} on the {Challenges} of {Abstractive} {Dialogue} {Summarization}},\n\tvolume = {82},\n\tissn = {1076-9757},\n\tshorttitle = {{CADS}},\n\turl = {http://jair.org/index.php/jair/article/view/16674},\n\tdoi = {10.1613/jair.1.16674},\n\tabstract = {Abstractive dialogue summarization is the task of distilling conversations into informative and concise summaries. Although focused reviews have been conducted on this topic, there is a lack of comprehensive work that details the core challenges of dialogue summarization, unifies the differing understanding of the task, and aligns proposed techniques, datasets, and evaluation metrics with the challenges. This article summarizes the research on Transformer-based abstractive summarization for English dialogues by systematically reviewing 1262 unique research papers published between 2019 and 2024, relying on the Semantic Scholar and DBLP databases. We cover the main challenges present in dialog summarization (i.e., language, structure, comprehension, speaker, salience, and factuality) and link them to corresponding techniques such as graph-based approaches, additional training tasks, and planning strategies, which typically overly rely on BART-based encoder-decoder models. Recent advances in training methods have led to substantial improvements in language-related challenges. However, challenges such as comprehension, factuality, and salience remain difficult and present significant research opportunities. We further investigate how these approaches are typically analyzed, covering the datasets for the subdomains of dialogue (e.g., meeting, customer service, and medical), the established automatic metrics (e.g., ROUGE), and common human evaluation approaches for assigning scores and evaluating annotator agreement. We observe that only a few datasets (i.e., SAMSum, AMI, DialogSum) are widely used. Despite its limitations, the ROUGE metric is the most commonly used, while human evaluation, considered the gold standard, is frequently reported without sufficient detail on the inter-annotator agreement and annotation guidelines. Additionally, we discuss the possible implications of the recently explored large language models and conclude that our described challenge taxonomy remains relevant despite a potential shift in relevance and difficulty.},\n\turldate = {2025-01-29},\n\tjournal = {Journal of Artificial Intelligence Research},\n\tauthor = {Kirstein, Frederic and Wahle, Jan Philip and Gipp, Bela and Ruas, Terry},\n\tmonth = jan,\n\tyear = {2025},\n\tpages = {313--365},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Abstractive dialogue summarization is the task of distilling conversations into informative and concise summaries. Although focused reviews have been conducted on this topic, there is a lack of comprehensive work that details the core challenges of dialogue summarization, unifies the differing understanding of the task, and aligns proposed techniques, datasets, and evaluation metrics with the challenges. This article summarizes the research on Transformer-based abstractive summarization for English dialogues by systematically reviewing 1262 unique research papers published between 2019 and 2024, relying on the Semantic Scholar and DBLP databases. We cover the main challenges present in dialog summarization (i.e., language, structure, comprehension, speaker, salience, and factuality) and link them to corresponding techniques such as graph-based approaches, additional training tasks, and planning strategies, which typically overly rely on BART-based encoder-decoder models. Recent advances in training methods have led to substantial improvements in language-related challenges. However, challenges such as comprehension, factuality, and salience remain difficult and present significant research opportunities. We further investigate how these approaches are typically analyzed, covering the datasets for the subdomains of dialogue (e.g., meeting, customer service, and medical), the established automatic metrics (e.g., ROUGE), and common human evaluation approaches for assigning scores and evaluating annotator agreement. We observe that only a few datasets (i.e., SAMSum, AMI, DialogSum) are widely used. Despite its limitations, the ROUGE metric is the most commonly used, while human evaluation, considered the gold standard, is frequently reported without sufficient detail on the inter-annotator agreement and annotation guidelines. Additionally, we discuss the possible implications of the recently explored large language models and conclude that our described challenge taxonomy remains relevant despite a potential shift in relevance and difficulty.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2024\n \n \n (8)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n The Promises and Pitfalls of Using Language Models to Measure Instruction Quality in Education.\n \n \n \n \n\n\n \n Xu, P.; Liu, J.; Jones, N.; Cohen, J.; and Ai, W.\n\n\n \n\n\n\n In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 4375–4389, Mexico City, Mexico, 2024. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"ThePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{xu_promises_2024,\n\taddress = {Mexico City, Mexico},\n\ttitle = {The {Promises} and {Pitfalls} of {Using} {Language} {Models} to {Measure} {Instruction} {Quality} in {Education}},\n\turl = {https://aclanthology.org/2024.naacl-long.246},\n\tdoi = {10.18653/v1/2024.naacl-long.246},\n\tlanguage = {en},\n\turldate = {2025-06-25},\n\tbooktitle = {Proceedings of the 2024 {Conference} of the {North} {American} {Chapter} of the {Association} for {Computational} {Linguistics}: {Human} {Language} {Technologies} ({Volume} 1: {Long} {Papers})},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Xu, Paiheng and Liu, Jing and Jones, Nathan and Cohen, Julie and Ai, Wei},\n\tyear = {2024},\n\tkeywords = {!tr\\_author, media\\_bias},\n\tpages = {4375--4389},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Tell me what I need to know: Exploring LLM-based (Personalized) Abstractive Multi-Source Meeting Summarization.\n \n \n \n \n\n\n \n Kirstein, F.; Ruas, T.; Kratel, R.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 920–939, Miami, Florida, US, 2024. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"TellPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{kirstein_tell_2024,\n\taddress = {Miami, Florida, US},\n\ttitle = {Tell me what {I} need to know: {Exploring} {LLM}-based ({Personalized}) {Abstractive} {Multi}-{Source} {Meeting} {Summarization}},\n\tshorttitle = {Tell me what {I} need to know},\n\turl = {https://aclanthology.org/2024.emnlp-industry.69},\n\tdoi = {10.18653/v1/2024.emnlp-industry.69},\n\tlanguage = {en},\n\turldate = {2025-01-06},\n\tbooktitle = {Proceedings of the 2024 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}: {Industry} {Track}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kirstein, Frederic and Ruas, Terry and Kratel, Robert and Gipp, Bela},\n\tyear = {2024},\n\tpages = {920--939},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n What’s under the hood: Investigating Automatic Metrics on Meeting Summarization.\n \n \n \n \n\n\n \n Kirstein, F.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 6709–6723, Miami, Florida, USA, 2024. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"What’sPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{kirstein_whats_2024,\n\taddress = {Miami, Florida, USA},\n\ttitle = {What’s under the hood: {Investigating} {Automatic} {Metrics} on {Meeting} {Summarization}},\n\tshorttitle = {What’s under the hood},\n\turl = {https://aclanthology.org/2024.findings-emnlp.393},\n\tdoi = {10.18653/v1/2024.findings-emnlp.393},\n\tlanguage = {en},\n\turldate = {2025-01-06},\n\tbooktitle = {Findings of the {Association} for {Computational} {Linguistics}: {EMNLP} 2024},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kirstein, Frederic and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tyear = {2024},\n\tpages = {6709--6723},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Paraphrase Types Elicit Prompt Engineering Capabilities.\n \n \n \n \n\n\n \n Wahle, J. P.; Ruas, T.; Xu, Y.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 11004–11033, Miami, Florida, USA, 2024. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"ParaphrasePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{wahle_paraphrase_2024,\n\taddress = {Miami, Florida, USA},\n\ttitle = {Paraphrase {Types} {Elicit} {Prompt} {Engineering} {Capabilities}},\n\turl = {https://aclanthology.org/2024.emnlp-main.617},\n\tdoi = {10.18653/v1/2024.emnlp-main.617},\n\tlanguage = {en},\n\turldate = {2025-01-06},\n\tbooktitle = {Proceedings of the 2024 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Wahle, Jan Philip and Ruas, Terry and Xu, Yang and Gipp, Bela},\n\tyear = {2024},\n\tpages = {11004--11033},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n CiteAssist: A System for Automated Preprint Citation and BibTeX Generation.\n \n \n \n \n\n\n \n Kaesberg, L.; Ruas, T.; Wahle, J. P.; and Gipp, B.\n\n\n \n\n\n\n In Ghosal, T.; Singh, A.; Waard, A.; Mayr, P.; Naik, A.; Weller, O.; Lee, Y.; Shen, S.; and Qin, Y., editor(s), Proceedings of the Fourth Workshop on Scholarly Document Processing (SDP 2024), pages 105–119, Bangkok, Thailand, August 2024. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"CiteAssist:Paper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{kaesberg-etal-2024-citeassist,\n\taddress = {Bangkok, Thailand},\n\ttitle = {{CiteAssist}: {A} {System} for {Automated} {Preprint} {Citation} and {BibTeX} {Generation}},\n\turl = {https://aclanthology.org/2024.sdp-1.10},\n\tabstract = {We present CiteAssist, a system to automate the generation of BibTeX entries for preprints, streamlining the process of bibliographic annotation. Our system extracts metadata, such as author names, titles, publication dates, and keywords, to create standardized annotations within the document. CiteAssist automatically attaches the BibTeX citation to the end of a PDF and links it on the first page of the document so other researchers gain immediate access to the correct citation of the article. This method promotes platform flexibility by ensuring that annotations remain accessible regardless of the repository used to publish or access the preprint. The annotations remain available even if the preprint is viewed externally to CiteAssist. Additionally, the system adds relevant related papers based on extracted keywords to the preprint, providing researchers with additional publications besides those in related work for further reading. Researchers can enhance their preprints organization and reference management workflows through a free and publicly available web interface.},\n\tbooktitle = {Proceedings of the {Fourth} {Workshop} on {Scholarly} {Document} {Processing} ({SDP} 2024)},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kaesberg, Lars and Ruas, Terry and Wahle, Jan Philip and Gipp, Bela},\n\teditor = {Ghosal, Tirthankar and Singh, Amanpreet and Waard, Anita and Mayr, Philipp and Naik, Aakanksha and Weller, Orion and Lee, Yoonjoo and Shen, Shannon and Qin, Yanxia},\n\tmonth = aug,\n\tyear = {2024},\n\tkeywords = {!tr\\_author, tool, ws},\n\tpages = {105--119},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n We present CiteAssist, a system to automate the generation of BibTeX entries for preprints, streamlining the process of bibliographic annotation. Our system extracts metadata, such as author names, titles, publication dates, and keywords, to create standardized annotations within the document. CiteAssist automatically attaches the BibTeX citation to the end of a PDF and links it on the first page of the document so other researchers gain immediate access to the correct citation of the article. This method promotes platform flexibility by ensuring that annotations remain accessible regardless of the repository used to publish or access the preprint. The annotations remain available even if the preprint is viewed externally to CiteAssist. Additionally, the system adds relevant related papers based on extracted keywords to the preprint, providing researchers with additional publications besides those in related work for further reading. Researchers can enhance their preprints organization and reference management workflows through a free and publicly available web interface.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n MAGPIE: Multi-Task Analysis of Media-Bias Generalization with Pre-Trained Identification of Expressions.\n \n \n \n \n\n\n \n Horych, T.; Wessel, M. P.; Wahle, J. P.; Ruas, T.; Waßmuth, J.; Greiner-Petter, A.; Aizawa, A.; Gipp, B.; and Spinde, T.\n\n\n \n\n\n\n In Calzolari, N.; Kan, M.; Hoste, V.; Lenci, A.; Sakti, S.; and Xue, N., editor(s), Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 10903–10920, Torino, Italia, May 2024. ELRA and ICCL\n \n\n\n\n
\n\n\n\n \n \n \"MAGPIE:Paper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{horych_magpie_2024,\n\taddress = {Torino, Italia},\n\ttitle = {{MAGPIE}: {Multi}-{Task} {Analysis} of {Media}-{Bias} {Generalization} with {Pre}-{Trained} {Identification} of {Expressions}},\n\turl = {https://aclanthology.org/2024.lrec-main.952},\n\tabstract = {Media bias detection poses a complex, multifaceted problem traditionally tackled using single-task models and small in-domain datasets, consequently lacking generalizability. To address this, we introduce MAGPIE, a large-scale multi-task pre-training approach explicitly tailored for media bias detection. To enable large-scale pre-training, we construct Large Bias Mixture (LBM), a compilation of 59 bias-related tasks. MAGPIE outperforms previous approaches in media bias detection on the Bias Annotation By Experts (BABE) dataset, with a relative improvement of 3.3\\% F1-score. Furthermore, using a RoBERTa encoder, we show that MAGPIE needs only 15\\% of fine-tuning steps compared to single-task approaches. We provide insight into task learning interference and show that sentiment analysis and emotion detection help learning of all other tasks, and scaling the number of tasks leads to the best results. MAGPIE confirms that MTL is a promising approach for addressing media bias detection, enhancing the accuracy and efficiency of existing models. Furthermore, LBM is the first available resource collection focused on media bias MTL.},\n\tbooktitle = {Proceedings of the 2024 {Joint} {International} {Conference} on {Computational} {Linguistics}, {Language} {Resources} and {Evaluation} ({LREC}-{COLING} 2024)},\n\tpublisher = {ELRA and ICCL},\n\tauthor = {Horych, Tomáš and Wessel, Martin Paul and Wahle, Jan Philip and Ruas, Terry and Waßmuth, Jerome and Greiner-Petter, André and Aizawa, Akiko and Gipp, Bela and Spinde, Timo},\n\teditor = {Calzolari, Nicoletta and Kan, Min-Yen and Hoste, Veronique and Lenci, Alessandro and Sakti, Sakriani and Xue, Nianwen},\n\tmonth = may,\n\tyear = {2024},\n\tpages = {10903--10920},\n}\n\n\n\n
\n
\n\n\n
\n Media bias detection poses a complex, multifaceted problem traditionally tackled using single-task models and small in-domain datasets, consequently lacking generalizability. To address this, we introduce MAGPIE, a large-scale multi-task pre-training approach explicitly tailored for media bias detection. To enable large-scale pre-training, we construct Large Bias Mixture (LBM), a compilation of 59 bias-related tasks. MAGPIE outperforms previous approaches in media bias detection on the Bias Annotation By Experts (BABE) dataset, with a relative improvement of 3.3% F1-score. Furthermore, using a RoBERTa encoder, we show that MAGPIE needs only 15% of fine-tuning steps compared to single-task approaches. We provide insight into task learning interference and show that sentiment analysis and emotion detection help learning of all other tasks, and scaling the number of tasks leads to the best results. MAGPIE confirms that MTL is a promising approach for addressing media bias detection, enhancing the accuracy and efficiency of existing models. Furthermore, LBM is the first available resource collection focused on media bias MTL.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Text Generation: A Systematic Literature Review of Tasks, Evaluation, and Challenges.\n \n \n \n \n\n\n \n Becker, J.; Wahle, J. P.; Gipp, B.; and Ruas, T.\n\n\n \n\n\n\n May 2024.\n arXiv:2405.15604 [cs]\n\n\n\n
\n\n\n\n \n \n \"TextPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 2 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{becker_text_2024,\n\ttitle = {Text {Generation}: {A} {Systematic} {Literature} {Review} of {Tasks}, {Evaluation}, and {Challenges}},\n\tshorttitle = {Text {Generation}},\n\turl = {http://arxiv.org/abs/2405.15604},\n\tabstract = {Text generation has become more accessible than ever, and the increasing interest in these systems, especially those using large language models, has spurred an increasing number of related publications. We provide a systematic literature review comprising 244 selected papers between 2017 and 2024. This review categorizes works in text generation into five main tasks: open-ended text generation, summarization, translation, paraphrasing, and question answering. For each task, we review their relevant characteristics, sub-tasks, and specific challenges (e.g., missing datasets for multi-document summarization, coherence in story generation, and complex reasoning for question answering). Additionally, we assess current approaches for evaluating text generation systems and ascertain problems with current metrics. Our investigation shows nine prominent challenges common to all tasks and sub-tasks in recent text generation publications: bias, reasoning, hallucinations, misuse, privacy, interpretability, transparency, datasets, and computing. We provide a detailed analysis of these challenges, their potential solutions, and which gaps still require further engagement from the community. This systematic literature review targets two main audiences: early career researchers in natural language processing looking for an overview of the field and promising research directions, as well as experienced researchers seeking a detailed view of tasks, evaluation methodologies, open challenges, and recent mitigation strategies.},\n\turldate = {2024-05-27},\n\tpublisher = {arXiv},\n\tauthor = {Becker, Jonas and Wahle, Jan Philip and Gipp, Bela and Ruas, Terry},\n\tmonth = may,\n\tyear = {2024},\n\tnote = {arXiv:2405.15604 [cs]},\n\tkeywords = {!tr\\_author, A.1, Computer Science - Computation and Language, I.2.7, nlp\\_text generation},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Text generation has become more accessible than ever, and the increasing interest in these systems, especially those using large language models, has spurred an increasing number of related publications. We provide a systematic literature review comprising 244 selected papers between 2017 and 2024. This review categorizes works in text generation into five main tasks: open-ended text generation, summarization, translation, paraphrasing, and question answering. For each task, we review their relevant characteristics, sub-tasks, and specific challenges (e.g., missing datasets for multi-document summarization, coherence in story generation, and complex reasoning for question answering). Additionally, we assess current approaches for evaluating text generation systems and ascertain problems with current metrics. Our investigation shows nine prominent challenges common to all tasks and sub-tasks in recent text generation publications: bias, reasoning, hallucinations, misuse, privacy, interpretability, transparency, datasets, and computing. We provide a detailed analysis of these challenges, their potential solutions, and which gaps still require further engagement from the community. This systematic literature review targets two main audiences: early career researchers in natural language processing looking for an overview of the field and promising research directions, as well as experienced researchers seeking a detailed view of tasks, evaluation methodologies, open challenges, and recent mitigation strategies.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n The Media Bias Taxonomy: A Systematic Literature Review on the Forms and Automated Detection of Media Bias.\n \n \n \n \n\n\n \n Spinde, T.; Hinterreiter, S.; Haak, F.; Ruas, T.; Giese, H.; Meuschke, N.; and Gipp, B.\n\n\n \n\n\n\n January 2024.\n arXiv:2312.16148 [cs]\n\n\n\n
\n\n\n\n \n \n \"ThePaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 5 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{spinde_media_2024,\n\ttitle = {The {Media} {Bias} {Taxonomy}: {A} {Systematic} {Literature} {Review} on the {Forms} and {Automated} {Detection} of {Media} {Bias}},\n\tshorttitle = {The {Media} {Bias} {Taxonomy}},\n\turl = {http://arxiv.org/abs/2312.16148},\n\tabstract = {The way the media presents events can significantly affect public perception, which in turn can alter people's beliefs and views. Media bias describes a one-sided or polarizing perspective on a topic. This article summarizes the research on computational methods to detect media bias by systematically reviewing 3140 research papers published between 2019 and 2022. To structure our review and support a mutual understanding of bias across research domains, we introduce the Media Bias Taxonomy, which provides a coherent overview of the current state of research on media bias from different perspectives. We show that media bias detection is a highly active research field, in which transformer-based classification approaches have led to significant improvements in recent years. These improvements include higher classification accuracy and the ability to detect more fine-granular types of bias. However, we have identified a lack of interdisciplinarity in existing projects, and a need for more awareness of the various types of media bias to support methodologically thorough performance evaluations of media bias detection systems. Concluding from our analysis, we see the integration of recent machine learning advancements with reliable and diverse bias assessment strategies from other research areas as the most promising area for future research contributions in the field.},\n\turldate = {2024-02-02},\n\tpublisher = {arXiv},\n\tauthor = {Spinde, Timo and Hinterreiter, Smi and Haak, Fabian and Ruas, Terry and Giese, Helge and Meuschke, Norman and Gipp, Bela},\n\tmonth = jan,\n\tyear = {2024},\n\tnote = {arXiv:2312.16148 [cs]},\n\tkeywords = {!tr, !tr\\_author, Computer Science - Computation and Language, nlp\\_lit\\_rev, nlp\\_media\\_bias},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n The way the media presents events can significantly affect public perception, which in turn can alter people's beliefs and views. Media bias describes a one-sided or polarizing perspective on a topic. This article summarizes the research on computational methods to detect media bias by systematically reviewing 3140 research papers published between 2019 and 2022. To structure our review and support a mutual understanding of bias across research domains, we introduce the Media Bias Taxonomy, which provides a coherent overview of the current state of research on media bias from different perspectives. We show that media bias detection is a highly active research field, in which transformer-based classification approaches have led to significant improvements in recent years. These improvements include higher classification accuracy and the ability to detect more fine-granular types of bias. However, we have identified a lack of interdisciplinarity in existing projects, and a need for more awareness of the various types of media bias to support methodologically thorough performance evaluations of media bias detection systems. Concluding from our analysis, we see the integration of recent machine learning advancements with reliable and diverse bias assessment strategies from other research areas as the most promising area for future research contributions in the field.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2023\n \n \n (7)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n We are Who We Cite: Bridges of Influence Between Natural Language Processing and Other Academic Fields.\n \n \n \n \n\n\n \n Wahle, J.; Ruas, T.; Abdalla, M.; Gipp, B.; and Mohammad, S.\n\n\n \n\n\n\n In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 12896–12913, Singapore, 2023. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"WePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{wahle_we_2023,\n\taddress = {Singapore},\n\ttitle = {We are {Who} {We} {Cite}: {Bridges} of {Influence} {Between} {Natural} {Language} {Processing} and {Other} {Academic} {Fields}},\n\tshorttitle = {We are {Who} {We} {Cite}},\n\turl = {https://aclanthology.org/2023.emnlp-main.797},\n\tdoi = {10.18653/v1/2023.emnlp-main.797},\n\tlanguage = {en},\n\turldate = {2023-12-18},\n\tbooktitle = {Proceedings of the 2023 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Wahle, Jan and Ruas, Terry and Abdalla, Mohamed and Gipp, Bela and Mohammad, Saif},\n\tyear = {2023},\n\tkeywords = {!tr\\_author, nlp\\_scientometric},\n\tpages = {12896--12913},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Paraphrase Types for Generation and Detection.\n \n \n \n \n\n\n \n Wahle, J.; Gipp, B.; and Ruas, T.\n\n\n \n\n\n\n In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 12148–12164, Singapore, 2023. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"ParaphrasePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{wahle_paraphrase_2023,\n\taddress = {Singapore},\n\ttitle = {Paraphrase {Types} for {Generation} and {Detection}},\n\turl = {https://aclanthology.org/2023.emnlp-main.746},\n\tdoi = {10.18653/v1/2023.emnlp-main.746},\n\tlanguage = {en},\n\turldate = {2023-12-18},\n\tbooktitle = {Proceedings of the 2023 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Wahle, Jan and Gipp, Bela and Ruas, Terry},\n\tyear = {2023},\n\tkeywords = {!tr, !tr\\_author, nlp\\_paraphrase, nlp\\_paraphrase\\_types},\n\tpages = {12148--12164},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n AI Usage Cards: Responsibly Reporting AI-Generated Content.\n \n \n \n \n\n\n \n Wahle, J. P.; Ruas, T.; Mohammad, S. M.; Meuschke, N.; and Gipp, B.\n\n\n \n\n\n\n In 2023 ACM/IEEE Joint Conference on Digital Libraries (JCDL), pages 282–284, Santa Fe, NM, USA, June 2023. IEEE\n http://arxiv.org/abs/2303.03886\n\n\n\n
\n\n\n\n \n \n \"AIPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{wahle_ai_2023,\n\taddress = {Santa Fe, NM, USA},\n\ttitle = {{AI} {Usage} {Cards}: {Responsibly} {Reporting} {AI}-{Generated} {Content}},\n\tisbn = {9798350399318},\n\tshorttitle = {{AI} {Usage} {Cards}},\n\turl = {https://ieeexplore.ieee.org/document/10266234/},\n\tdoi = {10.1109/JCDL57899.2023.00060},\n\turldate = {2023-10-24},\n\tbooktitle = {2023 {ACM}/{IEEE} {Joint} {Conference} on {Digital} {Libraries} ({JCDL})},\n\tpublisher = {IEEE},\n\tauthor = {Wahle, Jan Philip and Ruas, Terry and Mohammad, Saif M. and Meuschke, Norman and Gipp, Bela},\n\tmonth = jun,\n\tyear = {2023},\n\tnote = {http://arxiv.org/abs/2303.03886},\n\tkeywords = {!tr\\_author, ai\\_ethics, nlp\\_ethics},\n\tpages = {282--284},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n What's in the News? Towards Identification of Bias by Commission, Omission, and Source Selection (COSS).\n \n \n \n \n\n\n \n Zhukova, A.; Ruas, T.; Hamborg, F.; Donnay, K.; and Gipp, B.\n\n\n \n\n\n\n In 2023 ACM/IEEE Joint Conference on Digital Libraries (JCDL), pages 258–259, Santa Fe, NM, USA, June 2023. IEEE\n \n\n\n\n
\n\n\n\n \n \n \"What'sPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{zhukova_whats_2023,\n\taddress = {Santa Fe, NM, USA},\n\ttitle = {What's in the {News}? {Towards} {Identification} of {Bias} by {Commission}, {Omission}, and {Source} {Selection} ({COSS})},\n\tisbn = {9798350399318},\n\tshorttitle = {What's in the {News}?},\n\turl = {https://ieeexplore.ieee.org/document/10266224/},\n\tdoi = {10.1109/JCDL57899.2023.00050},\n\turldate = {2023-10-24},\n\tbooktitle = {2023 {ACM}/{IEEE} {Joint} {Conference} on {Digital} {Libraries} ({JCDL})},\n\tpublisher = {IEEE},\n\tauthor = {Zhukova, Anastasia and Ruas, Terry and Hamborg, Felix and Donnay, Karsten and Gipp, Bela},\n\tmonth = jun,\n\tyear = {2023},\n\tkeywords = {!tr\\_author, nlp\\_coref, nlp\\_media\\_bias},\n\tpages = {258--259},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n The Elephant in the Room: Analyzing the Presence of Big Tech in Natural Language Processing Research.\n \n \n \n \n\n\n \n Abdalla, M.; Wahle, J. P.; Lima Ruas, T.; Névéol, A.; Ducel, F.; Mohammad, S.; and Fort, K.\n\n\n \n\n\n\n In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 13141–13160, Toronto, Canada, 2023. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"ThePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n\n\n\n
\n 
@inproceedings{abdalla_elephant_2023,\n\taddress = {Toronto, Canada},\n\ttitle = {The {Elephant} in the {Room}: {Analyzing} the {Presence} of {Big} {Tech} in {Natural} {Language} {Processing} {Research}},\n\tshorttitle = {The {Elephant} in the {Room}},\n\turl = {https://aclanthology.org/2023.acl-long.734},\n\tdoi = {10.18653/v1/2023.acl-long.734},\n\tlanguage = {en},\n\turldate = {2023-08-09},\n\tbooktitle = {Proceedings of the 61st {Annual} {Meeting} of the {Association} for {Computational} {Linguistics} ({Volume} 1: {Long} {Papers})},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Abdalla, Mohamed and Wahle, Jan Philip and Lima Ruas, Terry and Névéol, Aurélie and Ducel, Fanny and Mohammad, Saif and Fort, Karen},\n\tyear = {2023},\n\tkeywords = {!tr\\_author},\n\tpages = {13141--13160},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Introducing MBIB - The First Media Bias Identification Benchmark Task and Dataset Collection.\n \n \n \n \n\n\n \n Wessel, M.; Horych, T.; Ruas, T.; Aizawa, A.; Gipp, B.; and Spinde, T.\n\n\n \n\n\n\n In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 2765–2774, Taipei Taiwan, July 2023. ACM\n \n\n\n\n
\n\n\n\n \n \n \"IntroducingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{wessel_introducing_2023,\n\taddress = {Taipei Taiwan},\n\ttitle = {Introducing {MBIB} - {The} {First} {Media} {Bias} {Identification} {Benchmark} {Task} and {Dataset} {Collection}},\n\tisbn = {978-1-4503-9408-6},\n\turl = {https://dl.acm.org/doi/10.1145/3539618.3591882},\n\tdoi = {10.1145/3539618.3591882},\n\tlanguage = {en},\n\turldate = {2023-07-21},\n\tbooktitle = {Proceedings of the 46th {International} {ACM} {SIGIR} {Conference} on {Research} and {Development} in {Information} {Retrieval}},\n\tpublisher = {ACM},\n\tauthor = {Wessel, Martin and Horych, Tomás and Ruas, Terry and Aizawa, Akiko and Gipp, Bela and Spinde, Timo},\n\tmonth = jul,\n\tyear = {2023},\n\tpages = {2765--2774},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Paraphrase Detection: Human vs. Machine Content.\n \n \n \n \n\n\n \n Becker, J.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n March 2023.\n arXiv:2303.13989 [cs]\n\n\n\n
\n\n\n\n \n \n \"ParaphrasePaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{becker_paraphrase_2023,\n\ttitle = {Paraphrase {Detection}: {Human} vs. {Machine} {Content}},\n\tshorttitle = {Paraphrase {Detection}},\n\turl = {http://arxiv.org/abs/2303.13989},\n\tabstract = {The growing prominence of large language models, such as GPT-4 and ChatGPT, has led to increased concerns over academic integrity due to the potential for machine-generated content and paraphrasing. Although studies have explored the detection of human- and machine-paraphrased content, the comparison between these types of content remains underexplored. In this paper, we conduct a comprehensive analysis of various datasets commonly employed for paraphrase detection tasks and evaluate an array of detection methods. Our findings highlight the strengths and limitations of different detection methods in terms of performance on individual datasets, revealing a lack of suitable machine-generated datasets that can be aligned with human expectations. Our main finding is that human-authored paraphrases exceed machine-generated ones in terms of difficulty, diversity, and similarity implying that automatically generated texts are not yet on par with human-level performance. Transformers emerged as the most effective method across datasets with TF-IDF excelling on semantically diverse corpora. Additionally, we identify four datasets as the most diverse and challenging for paraphrase detection.},\n\turldate = {2023-04-25},\n\tpublisher = {arXiv},\n\tauthor = {Becker, Jonas and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tmonth = mar,\n\tyear = {2023},\n\tnote = {arXiv:2303.13989 [cs]},\n\tkeywords = {!tr\\_author, Computer Science - Artificial Intelligence, Computer Science - Computation and Language, nlp\\_paraphrase},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n The growing prominence of large language models, such as GPT-4 and ChatGPT, has led to increased concerns over academic integrity due to the potential for machine-generated content and paraphrasing. Although studies have explored the detection of human- and machine-paraphrased content, the comparison between these types of content remains underexplored. In this paper, we conduct a comprehensive analysis of various datasets commonly employed for paraphrase detection tasks and evaluate an array of detection methods. Our findings highlight the strengths and limitations of different detection methods in terms of performance on individual datasets, revealing a lack of suitable machine-generated datasets that can be aligned with human expectations. Our main finding is that human-authored paraphrases exceed machine-generated ones in terms of difficulty, diversity, and similarity implying that automatically generated texts are not yet on par with human-level performance. Transformers emerged as the most effective method across datasets with TF-IDF excelling on semantically diverse corpora. Additionally, we identify four datasets as the most diverse and challenging for paraphrase detection.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2022\n \n \n (9)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Analyzing Multi-Task Learning for Abstractive Text Summarization.\n \n \n \n \n\n\n \n Kirstein, F. T.; Wahle, J. P.; Ruas, T.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the 2nd Workshop on Natural Language Generation, Evaluation, and Metrics (GEM), pages 54–77, Abu Dhabi, United Arab Emirates (Hybrid), 2022. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"AnalyzingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{kirstein_analyzing_2022,\n\taddress = {Abu Dhabi, United Arab Emirates (Hybrid)},\n\ttitle = {Analyzing {Multi}-{Task} {Learning} for {Abstractive} {Text} {Summarization}},\n\turl = {https://aclanthology.org/2022.gem-1.5},\n\tdoi = {10.18653/v1/2022.gem-1.5},\n\tlanguage = {en},\n\turldate = {2023-08-09},\n\tbooktitle = {Proceedings of the 2nd {Workshop} on {Natural} {Language} {Generation}, {Evaluation}, and {Metrics} ({GEM})},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Kirstein, Frederic Thomas and Wahle, Jan Philip and Ruas, Terry and Gipp, Bela},\n\tyear = {2022},\n\tkeywords = {!tr, !tr\\_author, nlp\\_text\\_sum, ws},\n\tpages = {54--77},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n How Large Language Models are Transforming Machine-Paraphrase Plagiarism.\n \n \n \n \n\n\n \n Wahle, J. P.; Ruas, T.; Kirstein, F.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 952–963, Abu Dhabi, United Arab Emirates, 2022. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"HowPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n\n\n\n
\n 
@inproceedings{wahle_how_2022,\n\taddress = {Abu Dhabi, United Arab Emirates},\n\ttitle = {How {Large} {Language} {Models} are {Transforming} {Machine}-{Paraphrase} {Plagiarism}},\n\turl = {https://aclanthology.org/2022.emnlp-main.62},\n\tdoi = {10.18653/v1/2022.emnlp-main.62},\n\tlanguage = {en},\n\turldate = {2023-08-09},\n\tbooktitle = {Proceedings of the 2022 {Conference} on {Empirical} {Methods} in {Natural} {Language} {Processing}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Wahle, Jan Philip and Ruas, Terry and Kirstein, Frederic and Gipp, Bela},\n\tyear = {2022},\n\tkeywords = {!tr\\_author},\n\tpages = {952--963},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n A Domain-Adaptive Pre-Training Approach for Language Bias Detection in News.\n \n \n \n \n\n\n \n Krieger, J.; Spinde, T.; Ruas, T.; Kulshrestha, J.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the 22nd ACM/IEEE Joint Conference on Digital Libraries, of JCDL '22, New York, NY, USA, June 2022. Association for Computing Machinery\n Number of pages: 7 Place: Cologne, Germany tex.articleno: 3\n\n\n\n
\n\n\n\n \n \n \"APaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{krieger_domain-adaptive_2022,\n\taddress = {New York, NY, USA},\n\tseries = {{JCDL} '22},\n\ttitle = {A {Domain}-{Adaptive} {Pre}-{Training} {Approach} for {Language} {Bias} {Detection} in {News}},\n\tisbn = {978-1-4503-9345-4},\n\turl = {https://doi.org/10.1145/3529372.3530932},\n\tdoi = {10.1145/3529372.3530932},\n\tabstract = {Media bias is a multi-faceted construct influencing individual behavior and collective decision-making. Slanted news reporting is the result of one-sided and polarized writing which can occur in various forms. In this work, we focus on an important form of media bias, i.e. bias by word choice. Detecting biased word choices is a challenging task due to its linguistic complexity and the lack of representative gold-standard corpora. We present DA-RoBERTa, a new state-of-the-art transformer-based model adapted to the media bias domain which identifies sentence-level bias with an F1 score of 0.814. In addition, we also train, DA-BERT and DA-BART, two more transformer models adapted to the bias domain. Our proposed domain-adapted models outperform prior bias detection approaches on the same data.},\n\tbooktitle = {Proceedings of the 22nd {ACM}/{IEEE} {Joint} {Conference} on {Digital} {Libraries}},\n\tpublisher = {Association for Computing Machinery},\n\tauthor = {Krieger, Jan-David and Spinde, Timo and Ruas, Terry and Kulshrestha, Juhi and Gipp, Bela},\n\tmonth = jun,\n\tyear = {2022},\n\tnote = {Number of pages: 7\nPlace: Cologne, Germany\ntex.articleno: 3},\n\tkeywords = {domain adaptive, media bias, neural classification, news slant, text analysis},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Media bias is a multi-faceted construct influencing individual behavior and collective decision-making. Slanted news reporting is the result of one-sided and polarized writing which can occur in various forms. In this work, we focus on an important form of media bias, i.e. bias by word choice. Detecting biased word choices is a challenging task due to its linguistic complexity and the lack of representative gold-standard corpora. We present DA-RoBERTa, a new state-of-the-art transformer-based model adapted to the media bias domain which identifies sentence-level bias with an F1 score of 0.814. In addition, we also train, DA-BERT and DA-BART, two more transformer models adapted to the bias domain. Our proposed domain-adapted models outperform prior bias detection approaches on the same data.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Identifying Machine-Paraphrased Plagiarism.\n \n \n \n \n\n\n \n Wahle, J. P.; Ruas, T.; Foltýnek, T.; Meuschke, N.; and Gipp, B.\n\n\n \n\n\n\n In Smits, M., editor(s), Information for a Better World: Shaping the Global Future, pages 393–413, Cham, 2022. Springer International Publishing\n \n\n\n\n
\n\n\n\n \n \n \"IdentifyingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 19 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{wahle_identifying_2022,\n\taddress = {Cham},\n\ttitle = {Identifying {Machine}-{Paraphrased} {Plagiarism}},\n\tisbn = {978-3-030-96957-8},\n\turl = {https://arxiv.org/pdf/2103.11909.pdf},\n\tdoi = {10.1007/978-3-030-96957-8_34},\n\tabstract = {Employing paraphrasing tools to conceal plagiarized text is a severe threat to academic integrity. To enable the detection of machine-paraphrased text, we evaluate the effectiveness of five pre-trained word embedding models combined with machine learning classifiers and state-of-the-art neural language models. We analyze preprints of research papers, graduation theses, and Wikipedia articles, which we paraphrased using different configurations of the tools SpinBot and SpinnerChief. The best performing technique, Longformer, achieved an average F1 score of 80.99\\% (F1 = 99.68\\% for SpinBot and F1 = 71.64\\% for SpinnerChief cases), while human evaluators achieved F1 = 78.4\\% for SpinBot and F1 = 65.6\\% for SpinnerChief cases. We show that the automated classification alleviates shortcomings of widely-used text-matching systems, such as Turnitin and PlagScan.},\n\tbooktitle = {Information for a {Better} {World}: {Shaping} the {Global} {Future}},\n\tpublisher = {Springer International Publishing},\n\tauthor = {Wahle, Jan Philip and Ruas, Terry and Foltýnek, Tomáš and Meuschke, Norman and Gipp, Bela},\n\teditor = {Smits, Malte},\n\tyear = {2022},\n\tpages = {393--413},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Employing paraphrasing tools to conceal plagiarized text is a severe threat to academic integrity. To enable the detection of machine-paraphrased text, we evaluate the effectiveness of five pre-trained word embedding models combined with machine learning classifiers and state-of-the-art neural language models. We analyze preprints of research papers, graduation theses, and Wikipedia articles, which we paraphrased using different configurations of the tools SpinBot and SpinnerChief. The best performing technique, Longformer, achieved an average F1 score of 80.99% (F1 = 99.68% for SpinBot and F1 = 71.64% for SpinnerChief cases), while human evaluators achieved F1 = 78.4% for SpinBot and F1 = 65.6% for SpinnerChief cases. We show that the automated classification alleviates shortcomings of widely-used text-matching systems, such as Turnitin and PlagScan.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Specialized Document Embeddings for Aspect-Based Similarity of Research Papers.\n \n \n \n \n\n\n \n Ostendorff, M.; Blume, T.; Ruas, T.; Gipp, B.; and Rehm, G.\n\n\n \n\n\n\n In Proceedings of the 22nd ACM/IEEE Joint Conference on Digital Libraries, of JCDL '22, New York, NY, USA, 2022. Association for Computing Machinery\n Number of pages: 12 Place: Cologne, Germany tex.articleno: 7\n\n\n\n
\n\n\n\n \n \n \"SpecializedPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{ostendorff_specialized_2022,\n\taddress = {New York, NY, USA},\n\tseries = {{JCDL} '22},\n\ttitle = {Specialized {Document} {Embeddings} for {Aspect}-{Based} {Similarity} of {Research} {Papers}},\n\tisbn = {978-1-4503-9345-4},\n\turl = {https://doi.org/10.1145/3529372.3530912},\n\tdoi = {10.1145/3529372.3530912},\n\tabstract = {Document embeddings and similarity measures underpin content-based recommender systems, whereby a document is commonly represented as a single generic embedding. However, similarity computed on single vector representations provides only one perspective on document similarity that ignores which aspects make two documents alike. To address this limitation, aspect-based similarity measures have been developed using document segmentation or pairwise multi-class document classification. While segmentation harms the document coherence, the pairwise classification approach scales poorly to large scale corpora. In this paper, we treat aspect-based similarity as a classical vector similarity problem in aspect-specific embedding spaces. We represent a document not as a single generic embedding but as multiple specialized embeddings. Our approach avoids document segmentation and scales linearly w.r.t. the corpus size. In an empirical study, we use the Papers with Code corpus containing 157, 606 research papers and consider the task, method, and dataset of the respective research papers as their aspects. We compare and analyze three generic document embeddings, six specialized document embeddings and a pairwise classification baseline in the context of research paper recommendations. As generic document embeddings, we consider FastText, SciBERT, and SPECTER. To compute the specialized document embeddings, we compare three alternative methods inspired by retrofitting, fine-tuning, and Siamese networks. In our experiments, Siamese SciBERT achieved the highest scores. Additional analyses indicate an implicit bias of the generic document embeddings towards the dataset aspect and against the method aspect of each research paper. Our approach of aspect-based document embeddings mitigates potential risks arising from implicit biases by making them explicit. This can, for example, be used for more diverse and explainable recommendations.},\n\tbooktitle = {Proceedings of the 22nd {ACM}/{IEEE} {Joint} {Conference} on {Digital} {Libraries}},\n\tpublisher = {Association for Computing Machinery},\n\tauthor = {Ostendorff, Malte and Blume, Till and Ruas, Terry and Gipp, Bela and Rehm, Georg},\n\tyear = {2022},\n\tnote = {Number of pages: 12\nPlace: Cologne, Germany\ntex.articleno: 7},\n\tkeywords = {aspect-based similarity, content-based recommender systems, document embeddings, document similarity, papers with code},\n}\n\n\n\n
\n
\n\n\n
\n Document embeddings and similarity measures underpin content-based recommender systems, whereby a document is commonly represented as a single generic embedding. However, similarity computed on single vector representations provides only one perspective on document similarity that ignores which aspects make two documents alike. To address this limitation, aspect-based similarity measures have been developed using document segmentation or pairwise multi-class document classification. While segmentation harms the document coherence, the pairwise classification approach scales poorly to large scale corpora. In this paper, we treat aspect-based similarity as a classical vector similarity problem in aspect-specific embedding spaces. We represent a document not as a single generic embedding but as multiple specialized embeddings. Our approach avoids document segmentation and scales linearly w.r.t. the corpus size. In an empirical study, we use the Papers with Code corpus containing 157, 606 research papers and consider the task, method, and dataset of the respective research papers as their aspects. We compare and analyze three generic document embeddings, six specialized document embeddings and a pairwise classification baseline in the context of research paper recommendations. As generic document embeddings, we consider FastText, SciBERT, and SPECTER. To compute the specialized document embeddings, we compare three alternative methods inspired by retrofitting, fine-tuning, and Siamese networks. In our experiments, Siamese SciBERT achieved the highest scores. Additional analyses indicate an implicit bias of the generic document embeddings towards the dataset aspect and against the method aspect of each research paper. Our approach of aspect-based document embeddings mitigates potential risks arising from implicit biases by making them explicit. This can, for example, be used for more diverse and explainable recommendations.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Exploiting Transformer-Based Multitask Learning for the Detection of Media Bias in News Articles.\n \n \n \n \n\n\n \n Spinde, T.; Krieger, J.; Ruas, T.; Mitrović, J.; Götz-Hahn, F.; Aizawa, A.; and Gipp, B.\n\n\n \n\n\n\n In Smits, M., editor(s), Information for a Better World: Shaping the Global Future, volume 13192, pages 225–235, Cham, 2022. Springer International Publishing\n \n\n\n\n
\n\n\n\n \n \n \"ExploitingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{smits_exploiting_2022,\n\taddress = {Cham},\n\ttitle = {Exploiting {Transformer}-{Based} {Multitask} {Learning} for the {Detection} of {Media} {Bias} in {News} {Articles}},\n\tvolume = {13192},\n\tisbn = {978-3-030-96956-1 978-3-030-96957-8},\n\turl = {https://link.springer.com/10.1007/978-3-030-96957-8_20},\n\tdoi = {https://doi.org/10.1007/978-3-030-96957-8_20},\n\tlanguage = {en},\n\turldate = {2022-03-04},\n\tbooktitle = {Information for a {Better} {World}: {Shaping} the {Global} {Future}},\n\tpublisher = {Springer International Publishing},\n\tauthor = {Spinde, Timo and Krieger, Jan-David and Ruas, Terry and Mitrović, Jelena and Götz-Hahn, Franz and Aizawa, Akiko and Gipp, Bela},\n\teditor = {Smits, Malte},\n\tyear = {2022},\n\tkeywords = {!tr\\_author, media\\_bias, nlp},\n\tpages = {225--235},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Testing the Generalization of Neural Language Models for COVID-19 Misinformation Detection.\n \n \n \n \n\n\n \n Wahle, J. P.; Ashok, N.; Ruas, T.; Meuschke, N.; Ghosal, T.; and Gipp, B.\n\n\n \n\n\n\n In Smits, M., editor(s), Information for a Better World: Shaping the Global Future, volume 13192, pages 381–392. Springer International Publishing, Cham, 2022.\n Series Title: Lecture Notes in Computer Science\n\n\n\n
\n\n\n\n \n \n \"TestingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 5 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@incollection{smits_testing_2022,\n\taddress = {Cham},\n\ttitle = {Testing the {Generalization} of {Neural} {Language} {Models} for {COVID}-19 {Misinformation} {Detection}},\n\tvolume = {13192},\n\tisbn = {978-3-030-96956-1 978-3-030-96957-8},\n\turl = {https://link.springer.com/10.1007/978-3-030-96957-8_33},\n\tabstract = {A drastic rise in potentially life-threatening misinformation has been a by-product of the COVID-19 pandemic. Computational support to identify false information within the massive body of data on the topic is crucial to prevent harm. Researchers proposed many methods for flagging online misinformation related to COVID-19. However, these methods predominantly target specific content types (e.g., news) or platforms (e.g., Twitter). The methods’ capabilities to generalize were largely unclear so far. We evaluate fifteen Transformer-based models on five COVID-19 misinformation datasets that include social media posts, news articles, and scientific papers to fill this gap. We show tokenizers and models tailored to COVID-19 data do not provide a significant advantage over general-purpose ones. Our study provides a realistic assessment of models for detecting COVID-19 misinformation. We expect that evaluating a broad spectrum of datasets and models will benefit future research in developing misinformation detection systems.},\n\tlanguage = {en},\n\turldate = {2022-11-11},\n\tbooktitle = {Information for a {Better} {World}: {Shaping} the {Global} {Future}},\n\tpublisher = {Springer International Publishing},\n\tauthor = {Wahle, Jan Philip and Ashok, Nischal and Ruas, Terry and Meuschke, Norman and Ghosal, Tirthankar and Gipp, Bela},\n\teditor = {Smits, Malte},\n\tyear = {2022},\n\tdoi = {10.1007/978-3-030-96957-8_33},\n\tnote = {Series Title: Lecture Notes in Computer Science},\n\tpages = {381--392},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n A drastic rise in potentially life-threatening misinformation has been a by-product of the COVID-19 pandemic. Computational support to identify false information within the massive body of data on the topic is crucial to prevent harm. Researchers proposed many methods for flagging online misinformation related to COVID-19. However, these methods predominantly target specific content types (e.g., news) or platforms (e.g., Twitter). The methods’ capabilities to generalize were largely unclear so far. We evaluate fifteen Transformer-based models on five COVID-19 misinformation datasets that include social media posts, news articles, and scientific papers to fill this gap. We show tokenizers and models tailored to COVID-19 data do not provide a significant advantage over general-purpose ones. Our study provides a realistic assessment of models for detecting COVID-19 misinformation. We expect that evaluating a broad spectrum of datasets and models will benefit future research in developing misinformation detection systems.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n D3: A Massive Dataset of Scholarly Metadata for Analyzing the State of Computer Science Research.\n \n \n \n \n\n\n \n Wahle, J. P.; Ruas, T.; Mohammad, S.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 2642–2651, Marseille, France, June 2022. European Language Resources Association\n \n\n\n\n
\n\n\n\n \n \n \"D3:Paper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{wahle_d3_2022,\n\taddress = {Marseille, France},\n\ttitle = {D3: {A} {Massive} {Dataset} of {Scholarly} {Metadata} for {Analyzing} the {State} of {Computer} {Science} {Research}},\n\turl = {https://aclanthology.org/2022.lrec-1.283},\n\tabstract = {DBLP is the largest open-access repository of scientific articles on computer science and provides metadata associated with publications, authors, and venues. We retrieved more than 6 million publications from DBLP and extracted pertinent metadata (e.g., abstracts, author affiliations, citations) from the publication texts to create the DBLP Discovery Dataset (D3). D3 can be used to identify trends in research activity, productivity, focus, bias, accessibility, and impact of computer science research. We present an initial analysis focused on the volume of computer science research (e.g., number of papers, authors, research activity), trends in topics of interest, and citation patterns. Our findings show that computer science is a growing research field (15\\% annually), with an active and collaborative researcher community. While papers in recent years present more bibliographical entries in comparison to previous decades, the average number of citations has been declining. Investigating papers' abstracts reveals that recent topic trends are clearly reflected in D3. Finally, we list further applications of D3 and pose supplemental research questions. The D3 dataset, our findings, and source code are publicly available for research purposes.},\n\tbooktitle = {Proceedings of the {Thirteenth} {Language} {Resources} and {Evaluation} {Conference}},\n\tpublisher = {European Language Resources Association},\n\tauthor = {Wahle, Jan Philip and Ruas, Terry and Mohammad, Saif and Gipp, Bela},\n\tmonth = jun,\n\tyear = {2022},\n\tpages = {2642--2651},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n DBLP is the largest open-access repository of scientific articles on computer science and provides metadata associated with publications, authors, and venues. We retrieved more than 6 million publications from DBLP and extracted pertinent metadata (e.g., abstracts, author affiliations, citations) from the publication texts to create the DBLP Discovery Dataset (D3). D3 can be used to identify trends in research activity, productivity, focus, bias, accessibility, and impact of computer science research. We present an initial analysis focused on the volume of computer science research (e.g., number of papers, authors, research activity), trends in topics of interest, and citation patterns. Our findings show that computer science is a growing research field (15% annually), with an active and collaborative researcher community. While papers in recent years present more bibliographical entries in comparison to previous decades, the average number of citations has been declining. Investigating papers' abstracts reveals that recent topic trends are clearly reflected in D3. Finally, we list further applications of D3 and pose supplemental research questions. The D3 dataset, our findings, and source code are publicly available for research purposes.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n CS-Insights: A System for Analyzing Computer Science Research.\n \n \n \n \n\n\n \n Ruas, T.; Wahle, J. P.; Küll, L.; Mohammad, S. M.; and Gipp, B.\n\n\n \n\n\n\n October 2022.\n arXiv:2210.06878 [cs]\n\n\n\n
\n\n\n\n \n \n \"CS-Insights:Paper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@misc{ruas_cs-insights_2022,\n\ttitle = {{CS}-{Insights}: {A} {System} for {Analyzing} {Computer} {Science} {Research}},\n\tshorttitle = {{CS}-{Insights}},\n\turl = {http://arxiv.org/abs/2210.06878},\n\tabstract = {This paper presents CS-Insights, an interactive web application to analyze computer science publications from DBLP through multiple perspectives. The dedicated interfaces allow its users to identify trends in research activity, productivity, accessibility, author's productivity, venues' statistics, topics of interest, and the impact of computer science research on other fields. CS-Insightsis publicly available, and its modular architecture can be easily adapted to domains other than computer science.},\n\turldate = {2022-10-14},\n\tpublisher = {arXiv},\n\tauthor = {Ruas, Terry and Wahle, Jan Philip and Küll, Lennart and Mohammad, Saif M. and Gipp, Bela},\n\tmonth = oct,\n\tyear = {2022},\n\tnote = {arXiv:2210.06878 [cs]},\n\tkeywords = {!tr\\_author, Computer Science - Computation and Language, Computer Science - Digital Libraries, nlp\\_demo, nlp\\_scientometric},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n This paper presents CS-Insights, an interactive web application to analyze computer science publications from DBLP through multiple perspectives. The dedicated interfaces allow its users to identify trends in research activity, productivity, accessibility, author's productivity, venues' statistics, topics of interest, and the impact of computer science research on other fields. CS-Insightsis publicly available, and its modular architecture can be easily adapted to domains other than computer science.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2021\n \n \n (5)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Are Neural Language Models Good Plagiarists? A Benchmark for Neural Paraphrase Detection.\n \n \n \n \n\n\n \n Wahle, J. P.; Ruas, T.; Meuschke, N.; and Gipp, B.\n\n\n \n\n\n\n In 2021 ACM/IEEE Joint Conference on Digital Libraries (JCDL), pages 226–229, September 2021. \n arXiv:2103.12450 [cs]\n\n\n\n
\n\n\n\n \n \n \"ArePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 4 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{wahle_are_2021,\n\ttitle = {Are {Neural} {Language} {Models} {Good} {Plagiarists}? {A} {Benchmark} for {Neural} {Paraphrase} {Detection}},\n\tshorttitle = {Are {Neural} {Language} {Models} {Good} {Plagiarists}?},\n\turl = {https://doi.org/10.1109/JCDL52503.2021.00065},\n\tdoi = {10.1109/JCDL52503.2021.00065},\n\tabstract = {The rise of language models such as BERT allows for high-quality text paraphrasing. This is a problem to academic integrity, as it is difficult to differentiate between original and machine-generated content. We propose a benchmark consisting of paraphrased articles using recent language models relying on the Transformer architecture. Our contribution fosters future research of paraphrase detection systems as it offers a large collection of aligned original and paraphrased documents, a study regarding its structure, classification experiments with state-of-the-art systems, and we make our findings publicly available.},\n\turldate = {2022-11-04},\n\tbooktitle = {2021 {ACM}/{IEEE} {Joint} {Conference} on {Digital} {Libraries} ({JCDL})},\n\tauthor = {Wahle, Jan Philip and Ruas, Terry and Meuschke, Norman and Gipp, Bela},\n\tmonth = sep,\n\tyear = {2021},\n\tnote = {arXiv:2103.12450 [cs]},\n\tkeywords = {Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Digital Libraries},\n\tpages = {226--229},\n}\n\n\n\n
\n
\n\n\n
\n The rise of language models such as BERT allows for high-quality text paraphrasing. This is a problem to academic integrity, as it is difficult to differentiate between original and machine-generated content. We propose a benchmark consisting of paraphrased articles using recent language models relying on the Transformer architecture. Our contribution fosters future research of paraphrase detection systems as it offers a large collection of aligned original and paraphrased documents, a study regarding its structure, classification experiments with state-of-the-art systems, and we make our findings publicly available.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Evaluating document representations for content-based legal literature recommendations.\n \n \n \n \n\n\n \n Ostendorff, M.; Ash, E.; Ruas, T.; Gipp, B.; Moreno-Schneider, J.; and Rehm, G.\n\n\n \n\n\n\n In Proceedings of the Eighteenth International Conference on Artificial Intelligence and Law, pages 109–118, São Paulo Brazil, June 2021. ACM\n \n\n\n\n
\n\n\n\n \n \n \"EvaluatingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{ostendorff_evaluating_2021,\n\taddress = {São Paulo Brazil},\n\ttitle = {Evaluating document representations for content-based legal literature recommendations},\n\tisbn = {978-1-4503-8526-8},\n\turl = {https://dl.acm.org/doi/10.1145/3462757.3466073},\n\tdoi = {10.1145/3462757.3466073},\n\tlanguage = {en},\n\turldate = {2024-01-22},\n\tbooktitle = {Proceedings of the {Eighteenth} {International} {Conference} on {Artificial} {Intelligence} and {Law}},\n\tpublisher = {ACM},\n\tauthor = {Ostendorff, Malte and Ash, Elliott and Ruas, Terry and Gipp, Bela and Moreno-Schneider, Julian and Rehm, Georg},\n\tmonth = jun,\n\tyear = {2021},\n\tpages = {109--118},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Detecting Cross-Language Plagiarism using Open Knowledge Graphs.\n \n \n \n \n\n\n \n Stegmüller, J.; Bauer-Marquart, F.; Meuschke, N.; Lima Ruas, T.; Schubotz, M.; and Gipp, B.\n\n\n \n\n\n\n In 2nd Workshop on Extraction and Evaluation of Knowledge Entities from Scientific Documents (EEKE2021) at JCDL2021, pages 10, September 2021. \n https://arxiv.org/abs/2111.09749\n\n\n\n
\n\n\n\n \n \n \"DetectingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{stegmuller_detecting_2021,\n\ttitle = {Detecting {Cross}-{Language} {Plagiarism} using {Open} {Knowledge} {Graphs}},\n\turl = {https://ceur-ws.org/Vol-3004/paper7.pdf},\n\tdoi = {https://doi.org/10.6084/m9.figshare.17212340.v3},\n\tbooktitle = {2nd {Workshop} on {Extraction} and {Evaluation} of {Knowledge} {Entities} from {Scientific} {Documents} ({EEKE2021}) at {JCDL2021}},\n\tauthor = {Stegmüller, Johannes and Bauer-Marquart, Fabian and Meuschke, Norman and Lima Ruas, Terry and Schubotz, Moritz and Gipp, Bela},\n\tmonth = sep,\n\tyear = {2021},\n\tnote = {https://arxiv.org/abs/2111.09749},\n\tkeywords = {80107 Natural Language Processing, FOS: Computer and information sciences},\n\tpages = {10},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Neural Media Bias Detection Using Distant Supervision With BABE - Bias Annotations By Experts.\n \n \n \n \n\n\n \n Spinde, T.; Plank, M.; Krieger, J.; Ruas, T.; Gipp, B.; and Aizawa, A.\n\n\n \n\n\n\n In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 1166–1177, Punta Cana, Dominican Republic, 2021. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"NeuralPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{spinde_neural_2021,\n\taddress = {Punta Cana, Dominican Republic},\n\ttitle = {Neural {Media} {Bias} {Detection} {Using} {Distant} {Supervision} {With} {BABE} - {Bias} {Annotations} {By} {Experts}},\n\turl = {https://aclanthology.org/2021.findings-emnlp.101},\n\tdoi = {10.18653/v1/2021.findings-emnlp.101},\n\tlanguage = {en},\n\turldate = {2022-01-07},\n\tbooktitle = {Findings of the {Association} for {Computational} {Linguistics}: {EMNLP} 2021},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Spinde, Timo and Plank, Manuel and Krieger, Jan-David and Ruas, Terry and Gipp, Bela and Aizawa, Akiko},\n\tyear = {2021},\n\tkeywords = {!tr\\_author, dataset, media\\_bias, nlp},\n\tpages = {1166--1177},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Incorporating Word Sense Disambiguation in Neural Language Models.\n \n \n \n \n\n\n \n Wahle, J. P.; Ruas, T.; Meuschke, N.; and Gipp, B.\n\n\n \n\n\n\n arXiv:2106.07967 [cs]. June 2021.\n arXiv: 2106.07967\n\n\n\n
\n\n\n\n \n \n \"IncorporatingPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@article{WahleRMG21a,\n\ttitle = {Incorporating {Word} {Sense} {Disambiguation} in {Neural} {Language} {Models}},\n\turl = {https://arxiv.org/pdf/2106.07967.pdf},\n\tabstract = {We present two supervised (pre-)training methods to incorporate gloss definitions from lexical resources into neural language models (LMs). The training improves our models' performance for Word Sense Disambiguation (WSD) but also benefits general language understanding tasks while adding almost no parameters. We evaluate our techniques with seven different neural LMs and find that XLNet is more suitable for WSD than BERT. Our best-performing methods exceeds state-of-the-art WSD techniques on the SemCor 3.0 dataset by 0.5\\% F1 and increase BERT's performance on the GLUE benchmark by 1.1\\% on average.},\n\turldate = {2021-06-19},\n\tjournal = {arXiv:2106.07967 [cs]},\n\tauthor = {Wahle, Jan Philip and Ruas, Terry and Meuschke, Norman and Gipp, Bela},\n\tmonth = jun,\n\tyear = {2021},\n\tnote = {arXiv: 2106.07967},\n\tkeywords = {!tr\\_author, Computer Science - Artificial Intelligence, Computer Science - Computation and Language, \\_pre\\_print, lang\\_model, nlp, wsd, ⛔ No DOI found},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n We present two supervised (pre-)training methods to incorporate gloss definitions from lexical resources into neural language models (LMs). The training improves our models' performance for Word Sense Disambiguation (WSD) but also benefits general language understanding tasks while adding almost no parameters. We evaluate our techniques with seven different neural LMs and find that XLNet is more suitable for WSD than BERT. Our best-performing methods exceeds state-of-the-art WSD techniques on the SemCor 3.0 dataset by 0.5% F1 and increase BERT's performance on the GLUE benchmark by 1.1% on average.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2020\n \n \n (5)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Detecting Machine-Obfuscated Plagiarism.\n \n \n \n \n\n\n \n Foltýnek, T.; Ruas, T.; Scharpf, P.; Meuschke, N.; Schubotz, M.; Grosky, W.; and Gipp, B.\n\n\n \n\n\n\n In Sundqvist, A.; Berget, G.; Nolin, J.; and Skjerdingstad, K. I., editor(s), Sustainable Digital Communities, volume 12051, pages 816–827. Springer International Publishing, Cham, 2020.\n Series Title: Lecture Notes in Computer Science\n\n\n\n
\n\n\n\n \n \n \"DetectingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 5 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@incollection{sundqvist_detecting_2020,\n\taddress = {Cham},\n\ttitle = {Detecting {Machine}-{Obfuscated} {Plagiarism}},\n\tvolume = {12051},\n\tisbn = {978-3-030-43686-5 978-3-030-43687-2},\n\turl = {http://link.springer.com/10.1007/978-3-030-43687-2_68},\n\tlanguage = {en},\n\turldate = {2024-01-22},\n\tbooktitle = {Sustainable {Digital} {Communities}},\n\tpublisher = {Springer International Publishing},\n\tauthor = {Foltýnek, Tomáš and Ruas, Terry and Scharpf, Philipp and Meuschke, Norman and Schubotz, Moritz and Grosky, William and Gipp, Bela},\n\teditor = {Sundqvist, Anneli and Berget, Gerd and Nolin, Jan and Skjerdingstad, Kjell Ivar},\n\tyear = {2020},\n\tdoi = {10.1007/978-3-030-43687-2_68},\n\tnote = {Series Title: Lecture Notes in Computer Science},\n\tpages = {816--827},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Aspect-based Document Similarity for Research Papers.\n \n \n \n \n\n\n \n Ostendorff, M.; Ruas, T.; Blume, T.; Gipp, B.; and Rehm, G.\n\n\n \n\n\n\n In Proceedings of the 28th International Conference on Computational Linguistics, pages 6194–6206, Barcelona, Spain (Online), 2020. International Committee on Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"Aspect-basedPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{ostendorff_aspect-based_2020,\n\taddress = {Barcelona, Spain (Online)},\n\ttitle = {Aspect-based {Document} {Similarity} for {Research} {Papers}},\n\turl = {https://www.aclweb.org/anthology/2020.coling-main.545},\n\tdoi = {10.18653/v1/2020.coling-main.545},\n\tlanguage = {en},\n\turldate = {2024-01-22},\n\tbooktitle = {Proceedings of the 28th {International} {Conference} on {Computational} {Linguistics}},\n\tpublisher = {International Committee on Computational Linguistics},\n\tauthor = {Ostendorff, Malte and Ruas, Terry and Blume, Till and Gipp, Bela and Rehm, Georg},\n\tyear = {2020},\n\tpages = {6194--6206},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Pairwise Multi-Class Document Classification for Semantic Relations between Wikipedia Articles.\n \n \n \n \n\n\n \n Ostendorff, M.; Ruas, T.; Schubotz, M.; Rehm, G.; and Gipp, B.\n\n\n \n\n\n\n In Proceedings of the ACM/IEEE Joint Conference on Digital Libraries in 2020, pages 127–136, Virtual Event China, August 2020. ACM\n \n\n\n\n
\n\n\n\n \n \n \"PairwisePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{ostendorff_pairwise_2020,\n\taddress = {Virtual Event China},\n\ttitle = {Pairwise {Multi}-{Class} {Document} {Classification} for {Semantic} {Relations} between {Wikipedia} {Articles}},\n\tisbn = {978-1-4503-7585-6},\n\turl = {https://dl.acm.org/doi/10.1145/3383583.3398525},\n\tdoi = {10.1145/3383583.3398525},\n\tlanguage = {en},\n\turldate = {2024-01-22},\n\tbooktitle = {Proceedings of the {ACM}/{IEEE} {Joint} {Conference} on {Digital} {Libraries} in 2020},\n\tpublisher = {ACM},\n\tauthor = {Ostendorff, Malte and Ruas, Terry and Schubotz, Moritz and Rehm, Georg and Gipp, Bela},\n\tmonth = aug,\n\tyear = {2020},\n\tpages = {127--136},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Math-word embedding in math search and semantic extraction.\n \n \n \n \n\n\n \n Greiner-Petter, A.; Youssef, A.; Ruas, T.; Miller, B. R.; Schubotz, M.; Aizawa, A.; and Gipp, B.\n\n\n \n\n\n\n Scientometrics, 125(3): 3017–3046. December 2020.\n \n\n\n\n
\n\n\n\n \n \n \"Math-wordPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@article{GreinerPetterYRM20a,\n\ttitle = {Math-word embedding in math search and semantic extraction},\n\tvolume = {125},\n\tissn = {0138-9130, 1588-2861},\n\turl = {https://link.springer.com/article/10.1007/s11192-020-03502-9},\n\tdoi = {10.1007/s11192-020-03502-9},\n\tabstract = {Abstract\n            Word embedding, which represents individual words with semantically fixed-length vectors, has made it possible to successfully apply deep learning to natural language processing tasks such as semantic role-modeling, question answering, and machine translation. As math text consists of natural text, as well as math expressions that similarly exhibit linear correlation and contextual characteristics, word embedding techniques can also be applied to math documents. However, while mathematics is a precise and accurate science, it is usually expressed through imprecise and less accurate descriptions, contributing to the relative dearth of machine learning applications for information retrieval in this domain. Generally, mathematical documents communicate their knowledge with an ambiguous, context-dependent, and non-formal language. Given recent advances in word embedding, it is worthwhile to explore their use and effectiveness in math information retrieval tasks, such as math language processing and semantic knowledge extraction. In this paper, we explore math embedding by testing it on several different scenarios, namely, (1) math-term similarity, (2) analogy, (3) numerical concept-modeling based on the centroid of the keywords that characterize a concept, (4) math search using query expansions, and (5) semantic extraction, i.e., extracting descriptive phrases for math expressions. Due to the lack of benchmarks, our investigations were performed using the arXiv collection of STEM documents and carefully selected illustrations on the Digital Library of Mathematical Functions (DLMF: NIST digital library of mathematical functions. Release 1.0.20 of 2018-09-1, 2018). Our results show that math embedding holds much promise for similarity, analogy, and search tasks. However, we also observed the need for more robust math embedding approaches. Moreover, we explore and discuss fundamental issues that we believe thwart the progress in mathematical information retrieval in the direction of machine learning.},\n\tlanguage = {en},\n\tnumber = {3},\n\turldate = {2022-01-07},\n\tjournal = {Scientometrics},\n\tauthor = {Greiner-Petter, André and Youssef, Abdou and Ruas, Terry and Miller, Bruce R. and Schubotz, Moritz and Aizawa, Akiko and Gipp, Bela},\n\tmonth = dec,\n\tyear = {2020},\n\tkeywords = {!tr\\_author, embeddings, math, nlp},\n\tpages = {3017--3046},\n}\n\n\n\n
\n
\n\n\n
\n Abstract Word embedding, which represents individual words with semantically fixed-length vectors, has made it possible to successfully apply deep learning to natural language processing tasks such as semantic role-modeling, question answering, and machine translation. As math text consists of natural text, as well as math expressions that similarly exhibit linear correlation and contextual characteristics, word embedding techniques can also be applied to math documents. However, while mathematics is a precise and accurate science, it is usually expressed through imprecise and less accurate descriptions, contributing to the relative dearth of machine learning applications for information retrieval in this domain. Generally, mathematical documents communicate their knowledge with an ambiguous, context-dependent, and non-formal language. Given recent advances in word embedding, it is worthwhile to explore their use and effectiveness in math information retrieval tasks, such as math language processing and semantic knowledge extraction. In this paper, we explore math embedding by testing it on several different scenarios, namely, (1) math-term similarity, (2) analogy, (3) numerical concept-modeling based on the centroid of the keywords that characterize a concept, (4) math search using query expansions, and (5) semantic extraction, i.e., extracting descriptive phrases for math expressions. Due to the lack of benchmarks, our investigations were performed using the arXiv collection of STEM documents and carefully selected illustrations on the Digital Library of Mathematical Functions (DLMF: NIST digital library of mathematical functions. Release 1.0.20 of 2018-09-1, 2018). Our results show that math embedding holds much promise for similarity, analogy, and search tasks. However, we also observed the need for more robust math embedding approaches. Moreover, we explore and discuss fundamental issues that we believe thwart the progress in mathematical information retrieval in the direction of machine learning.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Enhanced word embeddings using multi-semantic representation through lexical chains.\n \n \n \n \n\n\n \n Ruas, T.; Ferreira, C. P. H.; Gorsky, W.; França, F. O.; and Medeiros, D. M. R.\n\n\n \n\n\n\n Information Sciences, 532: 16 –32. 2020.\n https://www.sciencedirect.com/science/article/pii/S0020025520303911\n\n\n\n
\n\n\n\n \n \n \"EnhancedPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@article{ruas_enhanced_2020,\n\ttitle = {Enhanced word embeddings using multi-semantic representation through lexical chains},\n\tvolume = {532},\n\tissn = {0020-0255},\n\turl = {https://arxiv.org/pdf/2101.09023.pdf},\n\tdoi = {10.1016/j.ins.2020.04.048},\n\tjournal = {Information Sciences},\n\tauthor = {Ruas, Terry and Ferreira, Charles P. H. and Gorsky, William and França, Fabrício O. and Medeiros, Débora M. R.},\n\tyear = {2020},\n\tnote = {https://www.sciencedirect.com/science/article/pii/S0020025520303911},\n\tkeywords = {!tr\\_author, doc\\_classification, lexical\\_chains},\n\tpages = {16 --32},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2019\n \n \n (3)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Why Machines Cannot Learn Mathematics, Yet.\n \n \n \n \n\n\n \n Greiner-Petter, A.; Ruas, T.; Schubotz, M.; Aizawa, A.; Grosky, W.; and Gipp, B.\n\n\n \n\n\n\n In arXiv:1905.08359 [cs], May 2019. \n \n\n\n\n
\n\n\n\n \n \n \"WhyPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 2 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{GreinerPetterRSA19a,\n\ttitle = {Why {Machines} {Cannot} {Learn} {Mathematics}, {Yet}},\n\turl = {http://ceur-ws.org/Vol-2414/paper14.pdf},\n\tabstract = {Nowadays, Machine Learning (ML) is seen as the universal solution to improve the effectiveness of information retrieval (IR) methods. However, while mathematics is a precise and accurate science, it is usually expressed by less accurate and imprecise descriptions, contributing to the relative dearth of machine learning applications for IR in this domain. Generally, mathematical documents communicate their knowledge with an ambiguous, context-dependent, and non-formal language. Given recent advances in ML, it seems canonical to apply ML techniques to represent and retrieve mathematics semantically. In this work, we apply popular text embedding techniques to the arXiv collection of STEM documents and explore how these are unable to properly understand mathematics from that corpus. In addition, we also investigate the missing aspects that would allow mathematics to be learned by computers.},\n\turldate = {2020-12-18},\n\tbooktitle = {{arXiv}:1905.08359 [cs]},\n\tauthor = {Greiner-Petter, André and Ruas, Terry and Schubotz, Moritz and Aizawa, Akiko and Grosky, William and Gipp, Bela},\n\tmonth = may,\n\tyear = {2019},\n\tkeywords = {!tr\\_author, Computer Science - Artificial Intelligence, Computer Science - Digital Libraries, Computer Science - Information Retrieval, machine\\_learning, math, nlp, ⛔ No DOI found},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Nowadays, Machine Learning (ML) is seen as the universal solution to improve the effectiveness of information retrieval (IR) methods. However, while mathematics is a precise and accurate science, it is usually expressed by less accurate and imprecise descriptions, contributing to the relative dearth of machine learning applications for IR in this domain. Generally, mathematical documents communicate their knowledge with an ambiguous, context-dependent, and non-formal language. Given recent advances in ML, it seems canonical to apply ML techniques to represent and retrieve mathematics semantically. In this work, we apply popular text embedding techniques to the arXiv collection of STEM documents and explore how these are unable to properly understand mathematics from that corpus. In addition, we also investigate the missing aspects that would allow mathematics to be learned by computers.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Data Science for Software Engineers.\n \n \n \n \n\n\n \n Grosky, W.; and Ruas, T.\n\n\n \n\n\n\n In Pressman, R.; and Maxim, B., editor(s), Software engineering: A practitioner's approach 9th edition, pages 629–638. MCGraw Hill, The address of the publisher, 9 edition, September 2019.\n \n\n\n\n
\n\n\n\n \n \n \"DataPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@incollection{GroskyR19,\n\taddress = {The address of the publisher},\n\tedition = {9},\n\ttitle = {Data {Science} for {Software} {Engineers}},\n\tisbn = {1-259-87297-1},\n\turl = {https://www.amazon.com/Software-Engineering-Practitioners-Roger-Pressman/dp/1259872971},\n\tlanguage = {English},\n\tbooktitle = {Software engineering: {A} practitioner's approach 9th edition},\n\tpublisher = {MCGraw Hill},\n\tauthor = {Grosky, William and Ruas, Terry},\n\teditor = {Pressman, Roger and Maxim, Bruce},\n\tmonth = sep,\n\tyear = {2019},\n\tkeywords = {!tr\\_author, data\\_science, sw\\_eng},\n\tpages = {629--638},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Multi-sense embeddings through a word sense disambiguation process.\n \n \n \n \n\n\n \n Ruas, T.; Gorsky, W.; and Aizawa, A.\n\n\n \n\n\n\n Expert Systems with Applications, 136: 288 – 303. 2019.\n http://www.sciencedirect.com/science/article/pii/S0957417419304269\n\n\n\n
\n\n\n\n \n \n \"Multi-sensePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@article{ruas_multi-sense_2019,\n\ttitle = {Multi-sense embeddings through a word sense disambiguation process},\n\tvolume = {136},\n\tissn = {0957-4174},\n\turl = {https://arxiv.org/pdf/2101.08700.pdf},\n\tdoi = {10.1016/j.eswa.2019.06.026},\n\tjournal = {Expert Systems with Applications},\n\tauthor = {Ruas, Terry and Gorsky, William and Aizawa, Akiko},\n\tyear = {2019},\n\tnote = {http://www.sciencedirect.com/science/article/pii/S0957417419304269},\n\tkeywords = {!tr\\_author, embeddings, nlp, wsd},\n\tpages = {288 -- 303},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2018\n \n \n (1)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Semantic Feature Structure Extraction From Documents Based on Extended Lexical Chains.\n \n \n \n \n\n\n \n Ruas, T.; and Grosky, W.\n\n\n \n\n\n\n In Proceedings of the 9th Global Wordnet Conference, pages 87–96, Nanyang Technological University (NTU), Singapore, January 2018. Global Wordnet Association\n \n\n\n\n
\n\n\n\n \n \n \"SemanticPaper\n  \n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{ruas_semantic_2018,\n\taddress = {Nanyang Technological University (NTU), Singapore},\n\ttitle = {Semantic {Feature} {Structure} {Extraction} {From} {Documents} {Based} on {Extended} {Lexical} {Chains}},\n\turl = {https://aclanthology.org/2018.gwc-1.11},\n\tabstract = {The meaning of a sentence in a document is more easily determined if its constituent words exhibit cohesion with respect to their individual semantics. This paper explores the degree of cohesion among a document's words using lexical chains as a semantic representation of its meaning. Using a combination of diverse types of lexical chains, we develop a text document representation that can be used for semantic document retrieval. For our approach, we develop two kinds of lexical chains: (i) a multilevel flexible chain representation of the extracted semantic values, which is used to construct a fixed segmentation of these chains and constituent words in the text; and (ii) a fixed lexical chain obtained directly from the initial semantic representation from a document. The extraction and processing of concepts is performed using WordNet as a lexical database. The segmentation then uses these lexical chains to model the dispersion of concepts in the document. Representing each document as a high-dimensional vector, we use spherical k-means clustering to demonstrate that our approach performs better than previous techniques.},\n\tbooktitle = {Proceedings of the 9th {Global} {Wordnet} {Conference}},\n\tpublisher = {Global Wordnet Association},\n\tauthor = {Ruas, Terry and Grosky, William},\n\tmonth = jan,\n\tyear = {2018},\n\tpages = {87--96},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n The meaning of a sentence in a document is more easily determined if its constituent words exhibit cohesion with respect to their individual semantics. This paper explores the degree of cohesion among a document's words using lexical chains as a semantic representation of its meaning. Using a combination of diverse types of lexical chains, we develop a text document representation that can be used for semantic document retrieval. For our approach, we develop two kinds of lexical chains: (i) a multilevel flexible chain representation of the extracted semantic values, which is used to construct a fixed segmentation of these chains and constituent words in the text; and (ii) a fixed lexical chain obtained directly from the initial semantic representation from a document. The extraction and processing of concepts is performed using WordNet as a lexical database. The segmentation then uses these lexical chains to model the dispersion of concepts in the document. Representing each document as a high-dimensional vector, we use spherical k-means clustering to demonstrate that our approach performs better than previous techniques.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2017\n \n \n (5)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Science, technology and innovation exploration in biophotonics through a scientometric approach.\n \n \n \n \n\n\n \n Ruas, T. L.; Pereira, L.; and Grosky, W. I.\n\n\n \n\n\n\n In 2017 IEEE International Conference on Smart Technologies and Management for Computing, Communication, Controls, Energy and Materials (ICSTM), pages 36–43, Chennai, India, August 2017. IEEE\n \n\n\n\n
\n\n\n\n \n \n \"Science,Paper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{ruas_science_2017,\n\taddress = {Chennai, India},\n\ttitle = {Science, technology and innovation exploration in biophotonics through a scientometric approach},\n\tisbn = {978-1-5090-5918-8 978-1-5090-5905-8},\n\turl = {http://ieeexplore.ieee.org/document/8089124/},\n\tdoi = {10.1109/ICSTM.2017.8089124},\n\turldate = {2024-01-22},\n\tbooktitle = {2017 {IEEE} {International} {Conference} on {Smart} {Technologies} and {Management} for {Computing}, {Communication}, {Controls}, {Energy} and {Materials} ({ICSTM})},\n\tpublisher = {IEEE},\n\tauthor = {Ruas, Terry L. and Pereira, Luciana and Grosky, William I.},\n\tmonth = aug,\n\tyear = {2017},\n\tpages = {36--43},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n Text similarity using multilevel fixed lexical chains.\n \n \n \n\n\n \n Ruas, T.; and Grosky, W.\n\n\n \n\n\n\n In Brazilian graduate student conference (BRASCON), Los Angeles, USA, 2017. \n ISSN: 2472-3894 Place: University of Southern California\n\n\n\n
\n\n\n\n \n\n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{RuasG17,\n\taddress = {Los Angeles, USA},\n\ttitle = {Text similarity using multilevel fixed lexical chains},\n\tbooktitle = {Brazilian graduate student conference ({BRASCON})},\n\tauthor = {Ruas, Terry and Grosky, William},\n\tyear = {2017},\n\tnote = {ISSN: 2472-3894\nPlace: University of Southern California},\n\tkeywords = {!tr\\_author, doc\\_classification, lexical\\_chains, nlp, ⛔ No DOI found},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n The continuing reinvention of content-based retrieval: Multimedia is not dead.\n \n \n \n \n\n\n \n Grosky, W. I.; and Ruas, T. L.\n\n\n \n\n\n\n IEEE MultiMedia, 24(1): 6–11. January 2017.\n \n\n\n\n
\n\n\n\n \n \n \"ThePaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@article{GroskyR17,\n\ttitle = {The continuing reinvention of content-based retrieval: {Multimedia} is not dead},\n\tvolume = {24},\n\tissn = {1070-986X},\n\turl = {https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7849103},\n\tdoi = {10.1109/mmul.2017.7},\n\tnumber = {1},\n\tjournal = {IEEE MultiMedia},\n\tauthor = {Grosky, William I. and Ruas, Terry L.},\n\tmonth = jan,\n\tyear = {2017},\n\tkeywords = {!tr\\_author, multimedia},\n\tpages = {6--11},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Exploring and expanding the use of lexical chains in information retrieval.\n \n \n \n \n\n\n \n Ruas, T.; and Grosky, W.\n\n\n \n\n\n\n Technical Report University of Michigan - Dearborn, Michigan, 2017.\n http://hdl.handle.net/2027.42/136659\n\n\n\n
\n\n\n\n \n \n \"ExploringPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@techreport{RuasG17b,\n\taddress = {Michigan},\n\ttitle = {Exploring and expanding the use of lexical chains in information retrieval},\n\turl = {https://deepblue.lib.umich.edu/bitstream/handle/2027.42/136659/LexicalChainsReport.pdf?sequence=1},\n\tinstitution = {University of Michigan - Dearborn},\n\tauthor = {Ruas, Terry and Grosky, William},\n\tyear = {2017},\n\tdoi = {10.3998/2027.42/136659},\n\tnote = {http://hdl.handle.net/2027.42/136659},\n\tkeywords = {!tr\\_author, ir, lexical\\_chains, nlp},\n\tpages = {6},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Keyword extraction through contextual semantic analysis of documents.\n \n \n \n \n\n\n \n Ruas, T.; and Grosky, W.\n\n\n \n\n\n\n In Proceedings of the 9th international conference on management of digital EcoSystems, of MEDES ’17, pages 150–156, New York, NY, USA, 2017. Association for Computing Machinery\n Place: Bangkok, Thailand\n\n\n\n
\n\n\n\n \n \n \"KeywordPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 2 downloads\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{RuasG17a,\n\taddress = {New York, NY, USA},\n\tseries = {{MEDES} ’17},\n\ttitle = {Keyword extraction through contextual semantic analysis of documents},\n\tisbn = {978-1-4503-4895-9},\n\turl = {https://doi.org/10.1145/3167020.3167043},\n\tdoi = {10.1145/3167020.3167043},\n\tbooktitle = {Proceedings of the 9th international conference on management of digital {EcoSystems}},\n\tpublisher = {Association for Computing Machinery},\n\tauthor = {Ruas, Terry and Grosky, William},\n\tyear = {2017},\n\tnote = {Place: Bangkok, Thailand},\n\tkeywords = {!tr\\_author, keyword\\_extraction, nlp},\n\tpages = {150--156},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2016\n \n \n (1)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Automated refactoring of ATL model transformations: A search-based approach.\n \n \n \n \n\n\n \n Alkhazi, B.; Ruas, T.; Kessentini, M.; Wimmer, M.; and Grosky, W. I.\n\n\n \n\n\n\n In Proceedings of the ACM/IEEE 19th international conference on model driven engineering languages and systems, of MODELS ’16, pages 295–304, New York, NY, USA, 2016. Association for Computing Machinery\n Place: Saint-malo, France\n\n\n\n
\n\n\n\n \n \n \"AutomatedPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{AlkhaziRKW16,\n\taddress = {New York, NY, USA},\n\tseries = {{MODELS} ’16},\n\ttitle = {Automated refactoring of {ATL} model transformations: {A} search-based approach},\n\tisbn = {978-1-4503-4321-3},\n\turl = {https://doi.org/10.1145/2976767.2976782},\n\tdoi = {10.1145/2976767.2976782},\n\tbooktitle = {Proceedings of the {ACM}/{IEEE} 19th international conference on model driven engineering languages and systems},\n\tpublisher = {Association for Computing Machinery},\n\tauthor = {Alkhazi, Bader and Ruas, Terry and Kessentini, Marouane and Wimmer, Manuel and Grosky, William I.},\n\tyear = {2016},\n\tnote = {Place: Saint-malo, France},\n\tkeywords = {!tr\\_authors, sw\\_eng},\n\tpages = {295--304},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2014\n \n \n (1)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Como construir indicadores de Ciência, Tecnologia e Inovação usando Web of Science, Derwent World Patent Index, Bibexcel e Pajek?.\n \n \n \n \n\n\n \n Ruas, T. L.; and Pereira, L.\n\n\n \n\n\n\n Perspectivas em Ciência da Informação, 19(3): 52–81. September 2014.\n \n\n\n\n
\n\n\n\n \n \n \"ComoPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@article{ruas_como_2014,\n\ttitle = {Como construir indicadores de {Ciência}, {Tecnologia} e {Inovação} usando {Web} of {Science}, {Derwent} {World} {Patent} {Index}, {Bibexcel} e {Pajek}?},\n\tvolume = {19},\n\tissn = {1413-9936},\n\turl = {http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1413-99362014000300004&lng=pt&tlng=pt},\n\tdoi = {10.1590/1981-5344/1678},\n\tabstract = {O objetivo deste artigo é apresentar um processo que explique como usar softwares livres para buscar, extrair (Bibexcel) e visualizar dados (Pajek) dos repositórios de publicações científicas (Web of Science) e tecnológicas (Derwent World Patent Index) para construção de indicadores de produtividade científica e tecnológica. Espera-se que este processo contribua como um guia metodológico para a realização de estudos cientométricos, de tal modo que permita aos pesquisadores e aos gestores de Ciência, Tecnologia e Inovação (CTI) sem conhecimentos avançados em computação a obter informações de forma mais prática, porém confiáveis, dos repositórios e que, a partir delas, possam construir indicadores e elaborar avaliações mais precisas de CTI.\n          , \n            The goal of this paper is to present a process to retrieve (Bibexcel), organize, and visualize (Pajek), information from data repositories for research publications (Web of Science) and technological efforts (Derwent World Patent Index) to build science and technology indicators. The process contributes as a methodological guide for scientometric studies in such a way that enables researchers and managers in Science, Technology and Innovation studies without advanced skills in computer science how to deal with large repositories in a more practical, yet reliable, way. The result is development of indicators that will allow more accurate assessments of science, technology and innovation.},\n\tnumber = {3},\n\turldate = {2024-01-23},\n\tjournal = {Perspectivas em Ciência da Informação},\n\tauthor = {Ruas, Terry Lima and Pereira, Luciana},\n\tmonth = sep,\n\tyear = {2014},\n\tpages = {52--81},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n O objetivo deste artigo é apresentar um processo que explique como usar softwares livres para buscar, extrair (Bibexcel) e visualizar dados (Pajek) dos repositórios de publicações científicas (Web of Science) e tecnológicas (Derwent World Patent Index) para construção de indicadores de produtividade científica e tecnológica. Espera-se que este processo contribua como um guia metodológico para a realização de estudos cientométricos, de tal modo que permita aos pesquisadores e aos gestores de Ciência, Tecnologia e Inovação (CTI) sem conhecimentos avançados em computação a obter informações de forma mais prática, porém confiáveis, dos repositórios e que, a partir delas, possam construir indicadores e elaborar avaliações mais precisas de CTI. , The goal of this paper is to present a process to retrieve (Bibexcel), organize, and visualize (Pajek), information from data repositories for research publications (Web of Science) and technological efforts (Derwent World Patent Index) to build science and technology indicators. The process contributes as a methodological guide for scientometric studies in such a way that enables researchers and managers in Science, Technology and Innovation studies without advanced skills in computer science how to deal with large repositories in a more practical, yet reliable, way. The result is development of indicators that will allow more accurate assessments of science, technology and innovation.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2013\n \n \n (1)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Simulating the Diffusion of Innovation Process a Multi Agent Approach.\n \n \n \n \n\n\n \n Noronha, E.; Lima Ruas, T.; Marietto, M. d. G. B.; Steinberger-Elias, M.; Botelho, W. T.; França, R. d. S.; and Soares, C.\n\n\n \n\n\n\n ,424158 Bytes. 2013.\n Artwork Size: 424158 Bytes Publisher: figshare\n\n\n\n
\n\n\n\n \n \n \"SimulatingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@article{noronha_simulating_2013,\n\ttitle = {Simulating the {Diffusion} of {Innovation} {Process} a {Multi} {Agent} {Approach}},\n\tcopyright = {Creative Commons Attribution 4.0 International},\n\turl = {https://figshare.com/articles/conference_contribution/Simulating_the_Diffusion_of_Innovation_Process_a_Multi_Agent_Approach/16825831/1},\n\tdoi = {10.6084/M9.FIGSHARE.16825831.V1},\n\tabstract = {Multi-agent based simulations provide tools andtechniques to observe and analyze the emergentbehavior that happens due to the interactions amongagents. These agents could represent the actors in a realsituation, or they could be used to test or to verifyhypothesis, theories and to perform experiments in acontrolled environment (e.g. virtual world). It is usefulin certain cases, for instance when the event and itsinner workings are hard to observe, which is common insocial simulations. The diffusion of innovation theory,presented by Everett Rogers, provides a classification ofthe individuals in a social system related to how long aninnovation takes to spread into the system. Also, itdescribes how people form clusters based on thehomophily concept. This paper brings two hypothesesfor the diffusion of innovation and puts them into testby a multi-agent based simulation, running in theSwarm multi-agent based framework. Rogers’ theory,as well as multi-agent based simulations, is brieflypresented so they become a background for thepresented model. Also, the simulation and its results areshown. Finally, conclusions and future works arediscussed.},\n\turldate = {2024-01-23},\n\tauthor = {Noronha, Emerson and Lima Ruas, Terry and Marietto, Maria das Graças Bruno and Steinberger-Elias, Margarethe and Botelho, Wagner Tanaka and França, Robson dos Santos and Soares, Camila},\n\tyear = {2013},\n\tnote = {Artwork Size: 424158 Bytes\nPublisher: figshare},\n\tkeywords = {80110 Simulation and Modelling, FOS: Computer and information sciences},\n\tpages = {424158 Bytes},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Multi-agent based simulations provide tools andtechniques to observe and analyze the emergentbehavior that happens due to the interactions amongagents. These agents could represent the actors in a realsituation, or they could be used to test or to verifyhypothesis, theories and to perform experiments in acontrolled environment (e.g. virtual world). It is usefulin certain cases, for instance when the event and itsinner workings are hard to observe, which is common insocial simulations. The diffusion of innovation theory,presented by Everett Rogers, provides a classification ofthe individuals in a social system related to how long aninnovation takes to spread into the system. Also, itdescribes how people form clusters based on thehomophily concept. This paper brings two hypothesesfor the diffusion of innovation and puts them into testby a multi-agent based simulation, running in theSwarm multi-agent based framework. Rogers’ theory,as well as multi-agent based simulations, is brieflypresented so they become a background for thepresented model. Also, the simulation and its results areshown. Finally, conclusions and future works arediscussed.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2012\n \n \n (1)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Multi-agent based simulation to explore the dynamics in the diffusion of innovation phenomenon.\n \n \n \n \n\n\n \n Noronha, E. A.; Marietto, M. G. B.; Born, M. S.; Botelho, W. T.; Ruas, T. L.; França, R.; and Soares, C.\n\n\n \n\n\n\n International Journal of Simulation: Systems, Science and Technology, 13(5): 22–32. 2012.\n \n\n\n\n
\n\n\n\n \n \n \"Multi-agentPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@article{NoronhaMBB12,\n\ttitle = {Multi-agent based simulation to explore the dynamics in the diffusion of innovation phenomenon},\n\tvolume = {13},\n\tissn = {1473-804x},\n\turl = {https://ijssst.info/Vol-13/No-5/paper4.pdf},\n\tdoi = {10.5013/IJSSST.a.13.05.04},\n\tnumber = {5},\n\tjournal = {International Journal of Simulation: Systems, Science and Technology},\n\tauthor = {Noronha, Emerson A. and Marietto, Maria G. B. and Born, Margarethe S. and Botelho, Wagner T. and Ruas, Terry L. and França, Robson and Soares, Camila},\n\tyear = {2012},\n\tkeywords = {!tr\\_author, mssa},\n\tpages = {22--32},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2011\n \n \n (2)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Principles of agent-oriented programming.\n \n \n \n \n\n\n \n Batista, A. F.; Marietto, M. G. B.; Botelho, W. T.; Kobayashi, G.; Passos, B. A.; Castro, S.; and Ruas, T. L.\n\n\n \n\n\n\n In Alkhateeb, F.; Al Maghayreh, E. A.; and Doush, I. A., editor(s), Multi-agent systems - modeling, control, programming, simulations and applications. IntechOpen, Rijeka, 2011.\n \n\n\n\n
\n\n\n\n \n \n \"PrinciplesPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@incollection{BatistaMBK11,\n\taddress = {Rijeka},\n\ttitle = {Principles of agent-oriented programming},\n\turl = {https://www.intechopen.com/chapters/14520},\n\tbooktitle = {Multi-agent systems - modeling, control, programming, simulations and applications},\n\tpublisher = {IntechOpen},\n\tauthor = {Batista, André F. and Marietto, Maria G. B. and Botelho, Wagner T. and Kobayashi, Guiou and Passos, Brunno A. and Castro, Sidney and Ruas, Terry L.},\n\teditor = {Alkhateeb, Faisal and Al Maghayreh, Eslam Al and Doush, Iyad Abu},\n\tyear = {2011},\n\tdoi = {10.5772/14248},\n\tkeywords = {!tr\\_author, mssa},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Modeling artificial life through multi-agent based simulation.\n \n \n \n \n\n\n \n Ruas, T. L.; Marietto, M. G. B.; Batista, A. F.; França, R. S.; Heideker, A.; Noronha, E. A.; and Silva, F. A.\n\n\n \n\n\n\n In Alkhateeb, F.; Al Maghayreh, E. A.; and Doush, I. A., editor(s), Multi-agent systems - modeling, control, programming, simulations and applications. IntechOpen, Rijeka, 2011.\n \n\n\n\n
\n\n\n\n \n \n \"ModelingPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@incollection{RuasMBF11,\n\taddress = {Rijeka},\n\ttitle = {Modeling artificial life through multi-agent based simulation},\n\turl = {https://cdn.intechopen.com/pdfs/14508/InTech-Modeling_artificial_life_through_multi_agent_based_simulation.pdf},\n\tbooktitle = {Multi-agent systems - modeling, control, programming, simulations and applications},\n\tpublisher = {IntechOpen},\n\tauthor = {Ruas, Terry L. and Marietto, Maria G. B. and Batista, André F. and França, Robson S. and Heideker, Alexandre and Noronha, Emerson A. and Silva, Fábio A.},\n\teditor = {Alkhateeb, Faisal and Al Maghayreh, Eslam Al and Doush, Iyad Abu},\n\tyear = {2011},\n\tdoi = {10.5772/14313},\n\tkeywords = {!tr\\_author, mass},\n}\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2010\n \n \n (1)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n An agent-based model for the spread of the dengue fever: A swarm platform simulation approach.\n \n \n \n \n\n\n \n Jacintho, L. F. O.; Batista, A. F. M.; Ruas, T. L.; Marietto, M. G. B.; and Silva, F. A.\n\n\n \n\n\n\n In Proceedings of the 2010 spring simulation multiconference, of SpringSim ’10, San Diego, CA, USA, 2010. Society for Computer Simulation International\n Place: Orlando, Florida https://doi.org/10.1145/1878537.1878540\n\n\n\n
\n\n\n\n \n \n \"AnPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@inproceedings{JacinthoBRM10,\n\taddress = {San Diego, CA, USA},\n\tseries = {{SpringSim} ’10},\n\ttitle = {An agent-based model for the spread of the dengue fever: {A} swarm platform simulation approach},\n\tisbn = {978-1-4503-0069-8},\n\turl = {https://figshare.com/articles/conference_contribution/An_agent-based_model_for_the_spread_of_the_Dengue_fever_a_swarm_platform_simulation_approach/16825912},\n\tdoi = {10.1145/1878537.1878540},\n\tbooktitle = {Proceedings of the 2010 spring simulation multiconference},\n\tpublisher = {Society for Computer Simulation International},\n\tauthor = {Jacintho, Luís F. O. and Batista, André F. M. and Ruas, Terry L. and Marietto, Maria G. B. and Silva, Fábio A.},\n\tyear = {2010},\n\tnote = {Place: Orlando, Florida\nhttps://doi.org/10.1145/1878537.1878540},\n\tkeywords = {!tr\\_author, health, mssa},\n}\n\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2009\n \n \n (3)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Multi-agent Systems in Modeling and Simulation of Fire Spread.\n \n \n \n \n\n\n \n Ruas, T.; Batista, A. F. d. M.; and Marietto, M. d. G. B.\n\n\n \n\n\n\n ,1705357 Bytes. 2009.\n Artwork Size: 1705357 Bytes Publisher: figshare\n\n\n\n
\n\n\n\n \n \n \"Multi-agentPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@article{ruas_multi-agent_2009,\n\ttitle = {Multi-agent {Systems} in {Modeling} and {Simulation} of {Fire} {Spread}},\n\tcopyright = {Creative Commons Attribution 4.0 International},\n\turl = {https://figshare.com/articles/conference_contribution/2009_-_Multi-agent_Systems_in_Modeling_and_Simulation_of_Fire_Spread_pdf/16825882/2},\n\tdoi = {10.6084/M9.FIGSHARE.16825882.V2},\n\tabstract = {This paper presents a conceptual study for the phenomenon of fire spread. The proposed study is based on simulations using the RoboCup Rescue environment. Using a socio-cognitive approach, it can be used in three dimensions: (i) assist in proposing new structures or alternatives to situations of collective panic, checking the viability of their existence and operation; (ii) obtain a better understanding of the reasons social, anthropological, psychological, etc. that subsidize and direct the type of collective behavior in crowd panic; (iii) assist in decision making aimed at minimizing the loss of life in a situation of spreadingfire. One of the challenges to be overcome in this research is the integration of the different theories andcomputational environments in order to open paths for future researches. Making possible a further studyon phenomenon of panic in crowds in danger situation.},\n\turldate = {2024-01-23},\n\tauthor = {Ruas, Terry and Batista, André Filipe de Moraes and Marietto, Maria das Graças Bruno},\n\tyear = {2009},\n\tnote = {Artwork Size: 1705357 Bytes\nPublisher: figshare},\n\tkeywords = {80110 Simulation and Modelling, FOS: Computer and information sciences},\n\tpages = {1705357 Bytes},\n}\n\n\n\n
\n
\n\n\n
\n This paper presents a conceptual study for the phenomenon of fire spread. The proposed study is based on simulations using the RoboCup Rescue environment. Using a socio-cognitive approach, it can be used in three dimensions: (i) assist in proposing new structures or alternatives to situations of collective panic, checking the viability of their existence and operation; (ii) obtain a better understanding of the reasons social, anthropological, psychological, etc. that subsidize and direct the type of collective behavior in crowd panic; (iii) assist in decision making aimed at minimizing the loss of life in a situation of spreadingfire. One of the challenges to be overcome in this research is the integration of the different theories andcomputational environments in order to open paths for future researches. Making possible a further studyon phenomenon of panic in crowds in danger situation.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Simulação Multiagentes na Propagação de Incêndios: Uma Abordagem via Simuladores do Ambiente RoboCup Rescue.\n \n \n \n \n\n\n \n Ruas, T.; and Marietto, M. d. G. B.\n\n\n \n\n\n\n 2009.\n Pages: 477046 Bytes Publisher: figshare\n\n\n\n
\n\n\n\n \n \n \"SimulaçãoPaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n\n\n\n
\n 
@misc{ruas_simulacao_2009,\n\ttitle = {Simulação {Multiagentes} na {Propagação} de {Incêndios}: {Uma} {Abordagem} via {Simuladores} do {Ambiente} {RoboCup} {Rescue}},\n\tcopyright = {Creative Commons Attribution 4.0 International},\n\tshorttitle = {Simulação {Multiagentes} na {Propagação} de {Incêndios}},\n\turl = {https://figshare.com/articles/poster/Simula_o_Multiagentes_na_Propaga_o_de_Inc_ndios_Uma_Abordagem_via_Simuladores_do_Ambiente_RoboCup_Rescue/16825861/1},\n\tabstract = {O presente trabalho tem como objetivo estudar a propagação de incêndios em centros urbanos através do ambiente RoboCup Rescue (RCR) e seus simuladores, em uma abordagem multiagentes.},\n\turldate = {2024-01-23},\n\tauthor = {Ruas, Terry and Marietto, Maria das Graças Bruno},\n\tyear = {2009},\n\tdoi = {10.6084/M9.FIGSHARE.16825861.V1},\n\tnote = {Pages: 477046 Bytes\nPublisher: figshare},\n\tkeywords = {80110 Simulation and Modelling, FOS: Computer and information sciences},\n}\n\n\n\n
\n
\n\n\n
\n O presente trabalho tem como objetivo estudar a propagação de incêndios em centros urbanos através do ambiente RoboCup Rescue (RCR) e seus simuladores, em uma abordagem multiagentes.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n A Model for fire spreading by multi-agent systems: A RoboCup Rescue simulation and Swarm platform approach.\n \n \n \n \n\n\n \n Ruas, T. L.; Marietto, M. D. G. B.; Franca, R. D. S.; and Batista, A. F. D. M.\n\n\n \n\n\n\n In 2009 Second International Conference on the Applications of Digital Information and Web Technologies, pages 380–385, London, United Kingdom, August 2009. IEEE\n \n\n\n\n
\n\n\n\n \n \n \"APaper\n  \n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{ruas_model_2009,\n\taddress = {London, United Kingdom},\n\ttitle = {A {Model} for fire spreading by multi-agent systems: {A} {RoboCup} {Rescue} simulation and {Swarm} platform approach},\n\tisbn = {978-1-4244-4456-4},\n\tshorttitle = {A {Model} for fire spreading by multi-agent systems},\n\turl = {http://ieeexplore.ieee.org/document/5273874/},\n\tdoi = {10.1109/ICADIWT.2009.5273874},\n\turldate = {2024-01-23},\n\tbooktitle = {2009 {Second} {International} {Conference} on the {Applications} of {Digital} {Information} and {Web} {Technologies}},\n\tpublisher = {IEEE},\n\tauthor = {Ruas, Terry L. and Marietto, Maria Das Gracas B. and Franca, Robson Dos S. and Batista, Andre Filipe De M.},\n\tmonth = aug,\n\tyear = {2009},\n\tpages = {380--385},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n\n
\n\n\n\n\n\n
\n
\n\n\n\n\n
\n\n\n \n\n \n \n \n \n\n
\n"}; document.write(bibbase_data.data);