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%2Fchathurijayaweera%2F5270774&jsonp=1&jsonp=1\"></script>\n \n
\n\n PHP\n
\n \n <?php\n $contents = file_get_contents(\"https://bibbase.org/show?bib=https%3A%2F%2Fbibbase.org%2Fzotero-group%2Fchathurijayaweera%2F5270774&jsonp=1\");\n print_r($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%2Fchathurijayaweera%2F5270774&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 An Elicitation-Matrix Approach to Pragmatic Context Modeling in Low-Resource Machine Translation: The Case of Akuapem Twi.\n \n \n \n \n\n\n \n Yamoah, K. A.; Agyapong, G.; Scroggins, K.; Parekh, N.; Brinkley, D.; Jayaweera, C.; Gilda, S.; Dorr, B.; and Dorley, E.\n\n\n \n\n\n\n The International FLAIRS Conference Proceedings, 39(1). May 2026.\n \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 abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@article{yamoah_elicitation-matrix_2026,\n\ttitle = {An {Elicitation}-{Matrix} {Approach} to {Pragmatic} {Context} {Modeling} in {Low}-{Resource} {Machine} {Translation}: {The} {Case} of {Akuapem} {Twi}},\n\tvolume = {39},\n\tcopyright = {Copyright (c) 2026 Kweku Andoh Yamoah, Godfred Agyapong, Kevin Scroggins, Neel Parekh, Detravious Brinkley, Chathuri Jayaweera, Shlok Gilda, Bonnie Dorr, Emmanuel Dorley},\n\tissn = {2334-0762},\n\tshorttitle = {An {Elicitation}-{Matrix} {Approach} to {Pragmatic} {Context} {Modeling} in {Low}-{Resource} {Machine} {Translation}},\n\turl = {https://journals.flvc.org/FLAIRS/article/view/141846},\n\tdoi = {10.32473/flairs.39.1.141846},\n\tabstract = {Pragmatic ambiguity poses a major challenge for machine translation in low-resource languages like Akan, where a single English phrase may represent multiple pragmatic contexts and vice versa. To address this gap, we develop an elicitation matrix capturing key social and situational factors and use it to create a pragmatics‑focused Akan--English dataset of 863 annotated pairs. We then evaluate whether large language models (LLMs) can infer pragmatic context and whether explicit pragmatic tags improve translation selection choices. Across two models, three prompting strategies, and three experimental settings, human‑annotated pragmatic tags consistently yield the highest accuracy, with the largest gains on expansive (many‑to‑one) mappings. Chain‑of‑thought prompting further boosts performance. These findings indicate that pragmatic conditioning---rather than model size---is the primary driver of improvement, and they suggest that future models will benefit from incorporating pragmatic information during training and inference.},\n\tlanguage = {en},\n\tnumber = {1},\n\turldate = {2026-05-26},\n\tjournal = {The International FLAIRS Conference Proceedings},\n\tauthor = {Yamoah, Kweku Andoh and Agyapong, Godfred and Scroggins, Kevin and Parekh, Neel and Brinkley, Detravious and Jayaweera, Chathuri and Gilda, Shlok and Dorr, Bonnie and Dorley, Emmanuel},\n\tmonth = may,\n\tyear = {2026},\n}\n\n\n\n
\n
\n\n\n
\n Pragmatic ambiguity poses a major challenge for machine translation in low-resource languages like Akan, where a single English phrase may represent multiple pragmatic contexts and vice versa. To address this gap, we develop an elicitation matrix capturing key social and situational factors and use it to create a pragmatics‑focused Akan–English dataset of 863 annotated pairs. We then evaluate whether large language models (LLMs) can infer pragmatic context and whether explicit pragmatic tags improve translation selection choices. Across two models, three prompting strategies, and three experimental settings, human‑annotated pragmatic tags consistently yield the highest accuracy, with the largest gains on expansive (many‑to‑one) mappings. Chain‑of‑thought prompting further boosts performance. These findings indicate that pragmatic conditioning—rather than model size—is the primary driver of improvement, and they suggest that future models will benefit from incorporating pragmatic information during training and inference.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n AI You Can Trust: Communication-Aware, Ambiguity-Sensitive, and Interpretable NLP for High-Stakes Domains.\n \n \n \n \n\n\n \n Dorr, B. J.; Jayaweera, C.; Youm, S.; and Gilda, S.\n\n\n \n\n\n\n The International FLAIRS Conference Proceedings, 39(1). May 2026.\n \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 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 
@article{dorr_ai_2026,\n\ttitle = {{AI} {You} {Can} {Trust}: {Communication}-{Aware}, {Ambiguity}-{Sensitive}, and {Interpretable} {NLP} for {High}-{Stakes} {Domains}},\n\tvolume = {39},\n\tcopyright = {Copyright (c) 2026 Bonnie J. Dorr, Chathuri Jayaweera, Sangpil Youm, Shlok Gilda},\n\tissn = {2334-0762},\n\tshorttitle = {{AI} {You} {Can} {Trust}},\n\turl = {https://journals.flvc.org/FLAIRS/article/view/142076},\n\tdoi = {10.32473/flairs.39.1.142076},\n\tabstract = {Artificial intelligence systems increasingly interpret human communication in high-stakes settings such as mental health, legal reasoning, and cybersecurity contexts. Yet current large language models often produce fluent but incorrect outputs, especially when meaning depends on ambiguity, hidden mental states, or community-specific communication patterns. We argue that trustworthy AI in such settings requires a shift away from purely generative pipelines toward hybrid, communication-aware, structure-aware, and ambiguity-sensitive NLP that supports interpretable and reliable inference. We present a position supported by three complementary research directions: structure-aware analysis of mental health signals, ambiguity-aware reasoning for explainable inference in domains such as legal interpretation, and communication-driven risk modeling in open-source ecosystems. Across these case studies, we argue that reliable high-stakes AI must integrate linguistic and interaction structure, socio-communicative context, and explicit reasoning, while preserving human oversight and ethical safeguards.},\n\tlanguage = {en},\n\tnumber = {1},\n\turldate = {2026-05-26},\n\tjournal = {The International FLAIRS Conference Proceedings},\n\tauthor = {Dorr, Bonnie J. and Jayaweera, Chathuri and Youm, Sangpil and Gilda, Shlok},\n\tmonth = may,\n\tyear = {2026},\n\tkeywords = {Ambiguity-aware reasoning, High-stakes domains, Natural language processing, Neuro-symbolic AI, Semantic role labeling, Trustworthy AI},\n}\n\n\n\n
\n
\n\n\n
\n Artificial intelligence systems increasingly interpret human communication in high-stakes settings such as mental health, legal reasoning, and cybersecurity contexts. Yet current large language models often produce fluent but incorrect outputs, especially when meaning depends on ambiguity, hidden mental states, or community-specific communication patterns. We argue that trustworthy AI in such settings requires a shift away from purely generative pipelines toward hybrid, communication-aware, structure-aware, and ambiguity-sensitive NLP that supports interpretable and reliable inference. We present a position supported by three complementary research directions: structure-aware analysis of mental health signals, ambiguity-aware reasoning for explainable inference in domains such as legal interpretation, and communication-driven risk modeling in open-source ecosystems. Across these case studies, we argue that reliable high-stakes AI must integrate linguistic and interaction structure, socio-communicative context, and explicit reasoning, while preserving human oversight and ethical safeguards.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Filling the Gap: Is Commonsense Knowledge Generation useful for Natural Language Inference?.\n \n \n \n \n\n\n \n Jayaweera, C.; Yanqui, B.; and Dorr, B.\n\n\n \n\n\n\n January 2026.\n arXiv:2507.15100 [cs]\n\n\n\n
\n\n\n\n \n \n \"FillingPaper\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{jayaweera_filling_2026,\n\ttitle = {Filling the {Gap}: {Is} {Commonsense} {Knowledge} {Generation} useful for {Natural} {Language} {Inference}?},\n\tshorttitle = {Filling the {Gap}},\n\turl = {http://arxiv.org/abs/2507.15100},\n\tdoi = {10.48550/arXiv.2507.15100},\n\tabstract = {Natural Language Inference (NLI) is the task of determining whether a premise entails, contradicts, or is neutral with respect to a given hypothesis. The task is often framed as emulating human inferential processes, in which commonsense knowledge plays a major role. This study examines whether Large Language Models (LLMs) can generate useful commonsense axioms for Natural Language Inference, and evaluates their impact on performance using the SNLI and ANLI benchmarks with the Llama-3.1-70B and gpt-oss-120b models. We show that a hybrid approach, which selectively provides highly factual axioms based on judged helpfulness, yields consistent accuracy improvements of 1.99\\% to 6.88\\% across tested configurations, demonstrating the effectiveness of selective knowledge access for NLI. We also find that this targeted use of commonsense knowledge helps models overcome a bias toward the Neutral class by providing essential real-world context.},\n\turldate = {2026-02-24},\n\tpublisher = {arXiv},\n\tauthor = {Jayaweera, Chathuri and Yanqui, Brianna and Dorr, Bonnie},\n\tmonth = jan,\n\tyear = {2026},\n\tnote = {arXiv:2507.15100 [cs]},\n\tkeywords = {Computer Science - Artificial Intelligence, Computer Science - Computation and Language},\n}\n\n\n\n
\n
\n\n\n
\n Natural Language Inference (NLI) is the task of determining whether a premise entails, contradicts, or is neutral with respect to a given hypothesis. The task is often framed as emulating human inferential processes, in which commonsense knowledge plays a major role. This study examines whether Large Language Models (LLMs) can generate useful commonsense axioms for Natural Language Inference, and evaluates their impact on performance using the SNLI and ANLI benchmarks with the Llama-3.1-70B and gpt-oss-120b models. We show that a hybrid approach, which selectively provides highly factual axioms based on judged helpfulness, yields consistent accuracy improvements of 1.99% to 6.88% across tested configurations, demonstrating the effectiveness of selective knowledge access for NLI. We also find that this targeted use of commonsense knowledge helps models overcome a bias toward the Neutral class by providing essential real-world context.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2025\n \n \n (1)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n From Disagreement to Understanding: The Case for Ambiguity Detection in NLI.\n \n \n \n \n\n\n \n Jayaweera, C.; and Dorr, B. J.\n\n\n \n\n\n\n In Proceedings of the The 4th Workshop on Perspectivist Approaches to NLP, pages 37–46, Suzhou, China, November 2025. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"FromPaper\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{jayaweera_disagreement_2025,\n\taddress = {Suzhou, China},\n\ttitle = {From {Disagreement} to {Understanding}: {The} {Case} for {Ambiguity} {Detection} in {NLI}},\n\tisbn = {979-8-89176-350-0},\n\tshorttitle = {From {Disagreement} to {Understanding}},\n\turl = {https://aclanthology.org/2025.nlperspectives-1.4/},\n\tabstract = {This position paper argues that annotation disagreement in Natural Language Inference (NLI) is not mere noise but often reflects meaningful variation, especially when triggered by ambiguity in the premise or hypothesis. While underspecified guidelines and annotator behavior contribute to variation, content-based ambiguity provides a process-independent signal of divergent human perspectives. We call for a shift toward ambiguity-aware NLI that first identifies ambiguous input pairs, classifies their types, and only then proceeds to inference. To support this shift, we present a framework that incorporates ambiguity detection and classification prior to inference. We also introduce a unified taxonomy that synthesizes existing taxonomies, illustrates key subtypes with examples, and motivates targeted detection methods that better align models with human interpretation. Although current resources lack datasets explicitly annotated for ambiguity and subtypes, this gap presents an opportunity: by developing new annotated resources and exploring unsupervised approaches to ambiguity detection, we enable more robust, explainable, and human-aligned NLI systems.},\n\turldate = {2025-11-05},\n\tbooktitle = {Proceedings of the {The} 4th {Workshop} on {Perspectivist} {Approaches} to {NLP}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Jayaweera, Chathuri and Dorr, Bonnie J.},\n\tmonth = nov,\n\tyear = {2025},\n\tpages = {37--46},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n This position paper argues that annotation disagreement in Natural Language Inference (NLI) is not mere noise but often reflects meaningful variation, especially when triggered by ambiguity in the premise or hypothesis. While underspecified guidelines and annotator behavior contribute to variation, content-based ambiguity provides a process-independent signal of divergent human perspectives. We call for a shift toward ambiguity-aware NLI that first identifies ambiguous input pairs, classifies their types, and only then proceeds to inference. To support this shift, we present a framework that incorporates ambiguity detection and classification prior to inference. We also introduce a unified taxonomy that synthesizes existing taxonomies, illustrates key subtypes with examples, and motivates targeted detection methods that better align models with human interpretation. Although current resources lack datasets explicitly annotated for ambiguity and subtypes, this gap presents an opportunity: by developing new annotated resources and exploring unsupervised approaches to ambiguity detection, we enable more robust, explainable, and human-aligned NLI systems.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2024\n \n \n (3)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n AMREx: AMR for Explainable Fact Verification.\n \n \n \n \n\n\n \n Jayaweera, C.; Youm, S.; and Dorr, B. J\n\n\n \n\n\n\n In Proceedings of the Seventh Fact Extraction and VERification Workshop (FEVER), pages 234–244, Miami, Florida, USA, November 2024. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"AMREx: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 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{jayaweera_amrex_2024,\n\taddress = {Miami, Florida, USA},\n\ttitle = {{AMREx}: {AMR} for {Explainable} {Fact} {Verification}},\n\tshorttitle = {{AMREx}},\n\turl = {https://aclanthology.org/2024.fever-1.26},\n\tabstract = {With the advent of social media networks and the vast amount of information circulating through them, automatic fact verification is an essential component to prevent the spread of misinformation. It is even more useful to have fact verification systems that provide explanations along with their classifications to ensure accurate predictions. To address both of these requirements, we implement AMREx, an Abstract Meaning Representation (AMR)-based veracity prediction and explanation system for fact verification using a combination of Smatch, an AMR evaluation metric to measure meaning containment and textual similarity, and demonstrate its effectiveness in producing partially explainable justifications using two community standard fact verification datasets, FEVER and AVeriTeC. AMREx surpasses the AVeriTec baseline accuracy showing the effectiveness of our approach for real-world claim verification. It follows an interpretable pipeline and returns an explainable AMR node mapping to clarify the system's veracity predictions when applicable. We further demonstrate that AMREx output can be used to prompt LLMs to generate natural-language explanations using the AMR mappings as a guide to lessen the probability of hallucinations.},\n\turldate = {2024-11-25},\n\tbooktitle = {Proceedings of the {Seventh} {Fact} {Extraction} and {VERification} {Workshop} ({FEVER})},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Jayaweera, Chathuri and Youm, Sangpil and Dorr, Bonnie J},\n\tmonth = nov,\n\tyear = {2024},\n\tpages = {234--244},\n}\n\n\n\n
\n
\n\n\n
\n With the advent of social media networks and the vast amount of information circulating through them, automatic fact verification is an essential component to prevent the spread of misinformation. It is even more useful to have fact verification systems that provide explanations along with their classifications to ensure accurate predictions. To address both of these requirements, we implement AMREx, an Abstract Meaning Representation (AMR)-based veracity prediction and explanation system for fact verification using a combination of Smatch, an AMR evaluation metric to measure meaning containment and textual similarity, and demonstrate its effectiveness in producing partially explainable justifications using two community standard fact verification datasets, FEVER and AVeriTeC. AMREx surpasses the AVeriTec baseline accuracy showing the effectiveness of our approach for real-world claim verification. It follows an interpretable pipeline and returns an explainable AMR node mapping to clarify the system's veracity predictions when applicable. We further demonstrate that AMREx output can be used to prompt LLMs to generate natural-language explanations using the AMR mappings as a guide to lessen the probability of hallucinations.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Balancing Transparency and Accuracy: A Comparative Analysis of Rule-Based and Deep Learning Models in Political Bias Classification.\n \n \n \n \n\n\n \n Martinez, M. N.; Schmer-Galunder, S.; Liu, Z.; Youm, S.; Jayaweera, C.; and Dorr, B. J.\n\n\n \n\n\n\n In Proceedings of the Second Workshop on Social Influence in Conversations (SICon 2024), pages 102–115, Miami, Florida, USA, November 2024. Association for Computational Linguistics\n \n\n\n\n
\n\n\n\n \n \n \"BalancingPaper\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 
@inproceedings{martinez_balancing_2024,\n\taddress = {Miami, Florida, USA},\n\ttitle = {Balancing {Transparency} and {Accuracy}: {A} {Comparative} {Analysis} of {Rule}-{Based} and {Deep} {Learning} {Models} in {Political} {Bias} {Classification}},\n\tshorttitle = {Balancing {Transparency} and {Accuracy}},\n\turl = {https://aclanthology.org/2024.sicon-1.7},\n\tabstract = {The unchecked spread of digital information, combined with increasing political polarization and the tendency of individuals to isolate themselves from opposing political viewpoints opposing views, has driven researchers to develop systems for automatically detecting political bias in media. This trend has been further fueled by discussions on social media. We explore methods for categorizing bias in US news articles, comparing rule-based and deep learning approaches. The study highlights the sensitivity of modern self-learning systems to unconstrained data ingestion, while reconsidering the strengths of traditional rule-based systems. Applying both models to left-leaning (CNN) and right-leaning (FOX) News articles, we assess their effectiveness on data beyond the original training and test sets. This analysis highlights each model's accuracy, offers a framework for exploring deep-learning explainability, and sheds light on political bias in US news media. We contrast the opaque architecture of a deep learning model with the transparency of a linguistically informed rule-based model, showing that the rule-based model performs consistently across different data conditions and offers greater transparency, whereas the deep learning model is dependent on the training set and struggles with unseen data.},\n\turldate = {2024-11-25},\n\tbooktitle = {Proceedings of the {Second} {Workshop} on {Social} {Influence} in {Conversations} ({SICon} 2024)},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Martinez, Manuel Nunez and Schmer-Galunder, Sonja and Liu, Zoey and Youm, Sangpil and Jayaweera, Chathuri and Dorr, Bonnie J.},\n\tmonth = nov,\n\tyear = {2024},\n\tpages = {102--115},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n The unchecked spread of digital information, combined with increasing political polarization and the tendency of individuals to isolate themselves from opposing political viewpoints opposing views, has driven researchers to develop systems for automatically detecting political bias in media. This trend has been further fueled by discussions on social media. We explore methods for categorizing bias in US news articles, comparing rule-based and deep learning approaches. The study highlights the sensitivity of modern self-learning systems to unconstrained data ingestion, while reconsidering the strengths of traditional rule-based systems. Applying both models to left-leaning (CNN) and right-leaning (FOX) News articles, we assess their effectiveness on data beyond the original training and test sets. This analysis highlights each model's accuracy, offers a framework for exploring deep-learning explainability, and sheds light on political bias in US news media. We contrast the opaque architecture of a deep learning model with the transparency of a linguistically informed rule-based model, showing that the rule-based model performs consistently across different data conditions and offers greater transparency, whereas the deep learning model is dependent on the training set and struggles with unseen data.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n DAHRS: Divergence-Aware Hallucination-Remediated SRL Projection.\n \n \n \n\n\n \n Youm, S.; Mather, B.; Jayaweera, C.; Prada, J.; and Dorr, B.\n\n\n \n\n\n\n In Natural Language Processing and Information Systems, pages 423–438, Cham, 2024. Springer Nature Switzerland\n \n\n\n\n
\n\n\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 
@inproceedings{youm_dahrs_2024,\n\taddress = {Cham},\n\ttitle = {{DAHRS}: {Divergence}-{Aware} {Hallucination}-{Remediated} {SRL} {Projection}},\n\tisbn = {978-3-031-70239-6},\n\tshorttitle = {{DAHRS}},\n\tdoi = {10.1007/978-3-031-70239-6_29},\n\tabstract = {Semantic role labeling (SRL) enriches many downstream applications, e.g., machine translation, question answering, summarization, and stance/belief detection. However, building multilingual SRL models is challenging due to the scarcity of semantically annotated corpora for multiple languages. Moreover, state-of-the-art SRL projection (XSRL) based on large language models (LLMs) yields output that is riddled with spurious role labels. Remediation of such hallucinations is not straightforward due to the lack of explainability of LLMs. We show that hallucinated role labels are related to naturally occurring divergence types that interfere with initial alignments. We implement Divergence-Aware Hallucination-Remediated SRL projection (DAHRS), leveraging linguistically-informed alignment remediation followed by greedy First-Come First-Assign (FCFA) SRL projection. DAHRS improves the accuracy of SRL projection without additional transformer-based machinery, beating XSRL in both human and automatic comparisons, and advancing beyond headwords to accommodate phrase-level SRL projection (e.g., EN-FR, EN-ES). Using CoNLL-2009 as our ground truth, we achieve a higher word-level F1 over XSRL: 87.6\\% vs. 77.3\\% (EN-FR) and 89.0\\% vs. 82.7\\% (EN-ES). Human phrase-level assessments yield 89.1\\% (EN-FR) and 91.0\\% (EN-ES). We also define a divergence metric to adapt our approach to other language pairs (e.g., English-Tagalog).},\n\tlanguage = {en},\n\tbooktitle = {Natural {Language} {Processing} and {Information} {Systems}},\n\tpublisher = {Springer Nature Switzerland},\n\tauthor = {Youm, Sangpil and Mather, Brodie and Jayaweera, Chathuri and Prada, Juliana and Dorr, Bonnie},\n\tyear = {2024},\n\tpages = {423--438},\n}\n\n\n\n
\n
\n\n\n
\n Semantic role labeling (SRL) enriches many downstream applications, e.g., machine translation, question answering, summarization, and stance/belief detection. However, building multilingual SRL models is challenging due to the scarcity of semantically annotated corpora for multiple languages. Moreover, state-of-the-art SRL projection (XSRL) based on large language models (LLMs) yields output that is riddled with spurious role labels. Remediation of such hallucinations is not straightforward due to the lack of explainability of LLMs. We show that hallucinated role labels are related to naturally occurring divergence types that interfere with initial alignments. We implement Divergence-Aware Hallucination-Remediated SRL projection (DAHRS), leveraging linguistically-informed alignment remediation followed by greedy First-Come First-Assign (FCFA) SRL projection. DAHRS improves the accuracy of SRL projection without additional transformer-based machinery, beating XSRL in both human and automatic comparisons, and advancing beyond headwords to accommodate phrase-level SRL projection (e.g., EN-FR, EN-ES). Using CoNLL-2009 as our ground truth, we achieve a higher word-level F1 over XSRL: 87.6% vs. 77.3% (EN-FR) and 89.0% vs. 82.7% (EN-ES). Human phrase-level assessments yield 89.1% (EN-FR) and 91.0% (EN-ES). We also define a divergence metric to adapt our approach to other language pairs (e.g., English-Tagalog).\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2020\n \n \n (2)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n \n Automatic Classification, Visualization and Analysis of Errors in Machine Translation.\n \n \n \n \n\n\n \n Jayaweera, C.; and Dias, G.\n\n\n \n\n\n\n . 2020.\n Accepted: 2021-11-19T06:34:43Z\n\n\n\n
\n\n\n\n \n \n \"AutomaticPaper\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 
@article{jayaweera_automatic_2020,\n\ttitle = {Automatic {Classification}, {Visualization} and {Analysis} of {Errors} in {Machine} {Translation}},\n\turl = {http://repo.sltc.ac.lk/handle/456/131},\n\tabstract = {Although the quality of machine translation (MT) \nhas improved in recent years, machine translated documents \nstill contain errors. MT quality is often evaluated using a single \nnumeric score. However, this may not adequately characterise the \nsystem. We provide an error visualizer, which shows differences \nbetween corresponding lines of two translations. In addition to \ninsertions, deletions and substitutions, our system also shows \ntranspositions. We also provide an error analyzer which gives \nstatistics of each type of error in the document. In addition, it \nshows errors in context: the words commonly adjacent to each \nerror, and also the adjacent parts of speech (POS). This feature - \nunique to our system - allows the identification of the context in \nwhich errors occur, so they can be rectified easily. The system was \nevaluated by three MT system developers, who identified useful \nfeatures and provided feedback which was used to improve the \nsystem.},\n\tlanguage = {en},\n\turldate = {2023-11-06},\n\tpublisher = {Sri Lanka Technological Campus- IRC},\n\tauthor = {Jayaweera, Chathuri and Dias, Gihan},\n\tyear = {2020},\n\tnote = {Accepted: 2021-11-19T06:34:43Z},\n}\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Although the quality of machine translation (MT) has improved in recent years, machine translated documents still contain errors. MT quality is often evaluated using a single numeric score. However, this may not adequately characterise the system. We provide an error visualizer, which shows differences between corresponding lines of two translations. In addition to insertions, deletions and substitutions, our system also shows transpositions. We also provide an error analyzer which gives statistics of each type of error in the document. In addition, it shows errors in context: the words commonly adjacent to each error, and also the adjacent parts of speech (POS). This feature - unique to our system - allows the identification of the context in which errors occur, so they can be rectified easily. The system was evaluated by three MT system developers, who identified useful features and provided feedback which was used to improve the system.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n \n Vision-Based Adaptive Traffic Light Controller for Single Intersection.\n \n \n \n \n\n\n \n Sutharsan, M.; Rajakaruna, S.; Jayaweera, S.; Jayaweera, J.; and Thayaparan, S.\n\n\n \n\n\n\n In 2020 5th International Conference on Information Technology Research (ICITR), pages 1–6, December 2020. \n \n\n\n\n
\n\n\n\n \n \n \"Vision-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 abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{sutharsan_vision-based_2020,\n\ttitle = {Vision-{Based} {Adaptive} {Traffic} {Light} {Controller} for {Single} {Intersection}},\n\turl = {https://ieeexplore.ieee.org/abstract/document/9310872},\n\tdoi = {10.1109/ICITR51448.2020.9310872},\n\tabstract = {In this paper, a vision-based adaptive traffic light controller is proposed. The proposed controller was successfully deployed and tested as a complete system in a complex roundabout in Colombo city at a highly congested time. There were two main parts to this implementation. The first part was a vision-based traffic monitoring system. In this part, a system was developed so that it monitored lanes in a junction with cameras and extracted a traffic index based on traffic density, vehicle type, and pixel-wise velocity of vehicles by processing the video streams coming from cameras. The traffic signal light controlling part was the second part of the project. This part dealt with estimating a better timing adjustment for the existing system using a mathematical modeling approach while taking the extracted traffic index as input. This system was operated with the existing system with minimum alterations for easy real-world implementation. The developed prototype was plugged into the existing system to change traffic light phase timing according to the existing traffic level.},\n\turldate = {2023-11-06},\n\tbooktitle = {2020 5th {International} {Conference} on {Information} {Technology} {Research} ({ICITR})},\n\tauthor = {Sutharsan, Mahendren and Rajakaruna, Shehan and Jayaweera, S.Y. and Jayaweera, J.A.C.M. and Thayaparan, Subramaniam},\n\tmonth = dec,\n\tyear = {2020},\n\tpages = {1--6},\n}\n
\n
\n\n\n
\n In this paper, a vision-based adaptive traffic light controller is proposed. The proposed controller was successfully deployed and tested as a complete system in a complex roundabout in Colombo city at a highly congested time. There were two main parts to this implementation. The first part was a vision-based traffic monitoring system. In this part, a system was developed so that it monitored lanes in a junction with cameras and extracted a traffic index based on traffic density, vehicle type, and pixel-wise velocity of vehicles by processing the video streams coming from cameras. The traffic signal light controlling part was the second part of the project. This part dealt with estimating a better timing adjustment for the existing system using a mathematical modeling approach while taking the extracted traffic index as input. This system was operated with the existing system with minimum alterations for easy real-world implementation. The developed prototype was plugged into the existing system to change traffic light phase timing according to the existing traffic level.\n
\n\n\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);