A Survey on In-context Learning. Dong, Q., Li, L., Dai, D., Zheng, C., Wu, Z., Chang, B., Sun, X., Xu, J., Li, L., & Sui, Z. June, 2023. arXiv:2301.00234 [cs]![A Survey on In-context Learning [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex With the increasing ability of large language models (LLMs), in-context learning (ICL) has become a new paradigm for natural language processing (NLP), where LLMs make predictions only based on contexts augmented with a few examples. It has been a new trend to explore ICL to evaluate and extrapolate the ability of LLMs. In this paper, we aim to survey and summarize the progress and challenges of ICL. We first present a formal definition of ICL and clarify its correlation to related studies. Then, we organize and discuss advanced techniques, including training strategies, demonstration designing strategies, as well as related analysis. Finally, we discuss the challenges of ICL and provide potential directions for further research. We hope that our work can encourage more research on uncovering how ICL works and improving ICL.
@misc{dong_survey_2023,
title = {A {Survey} on {In}-context {Learning}},
url = {http://arxiv.org/abs/2301.00234},
abstract = {With the increasing ability of large language models (LLMs), in-context learning (ICL) has become a new paradigm for natural language processing (NLP), where LLMs make predictions only based on contexts augmented with a few examples. It has been a new trend to explore ICL to evaluate and extrapolate the ability of LLMs. In this paper, we aim to survey and summarize the progress and challenges of ICL. We first present a formal definition of ICL and clarify its correlation to related studies. Then, we organize and discuss advanced techniques, including training strategies, demonstration designing strategies, as well as related analysis. Finally, we discuss the challenges of ICL and provide potential directions for further research. We hope that our work can encourage more research on uncovering how ICL works and improving ICL.},
urldate = {2024-05-30},
publisher = {arXiv},
author = {Dong, Qingxiu and Li, Lei and Dai, Damai and Zheng, Ce and Wu, Zhiyong and Chang, Baobao and Sun, Xu and Xu, Jingjing and Li, Lei and Sui, Zhifang},
month = jun,
year = {2023},
note = {arXiv:2301.00234 [cs]},
keywords = {Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Jab/\#Pre},
}
Downloads: 0
{"_id":"G7aJ5yye3j2sMvxZX","bibbaseid":"dong-li-dai-zheng-wu-chang-sun-xu-etal-asurveyonincontextlearning-2023","author_short":["Dong, Q.","Li, L.","Dai, D.","Zheng, C.","Wu, Z.","Chang, B.","Sun, X.","Xu, J.","Li, L.","Sui, Z."],"bibdata":{"bibtype":"misc","type":"misc","title":"A Survey on In-context Learning","url":"http://arxiv.org/abs/2301.00234","abstract":"With the increasing ability of large language models (LLMs), in-context learning (ICL) has become a new paradigm for natural language processing (NLP), where LLMs make predictions only based on contexts augmented with a few examples. It has been a new trend to explore ICL to evaluate and extrapolate the ability of LLMs. In this paper, we aim to survey and summarize the progress and challenges of ICL. We first present a formal definition of ICL and clarify its correlation to related studies. Then, we organize and discuss advanced techniques, including training strategies, demonstration designing strategies, as well as related analysis. Finally, we discuss the challenges of ICL and provide potential directions for further research. We hope that our work can encourage more research on uncovering how ICL works and improving ICL.","urldate":"2024-05-30","publisher":"arXiv","author":[{"propositions":[],"lastnames":["Dong"],"firstnames":["Qingxiu"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["Lei"],"suffixes":[]},{"propositions":[],"lastnames":["Dai"],"firstnames":["Damai"],"suffixes":[]},{"propositions":[],"lastnames":["Zheng"],"firstnames":["Ce"],"suffixes":[]},{"propositions":[],"lastnames":["Wu"],"firstnames":["Zhiyong"],"suffixes":[]},{"propositions":[],"lastnames":["Chang"],"firstnames":["Baobao"],"suffixes":[]},{"propositions":[],"lastnames":["Sun"],"firstnames":["Xu"],"suffixes":[]},{"propositions":[],"lastnames":["Xu"],"firstnames":["Jingjing"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["Lei"],"suffixes":[]},{"propositions":[],"lastnames":["Sui"],"firstnames":["Zhifang"],"suffixes":[]}],"month":"June","year":"2023","note":"arXiv:2301.00234 [cs]","keywords":"Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Jab/#Pre","bibtex":"@misc{dong_survey_2023,\n\ttitle = {A {Survey} on {In}-context {Learning}},\n\turl = {http://arxiv.org/abs/2301.00234},\n\tabstract = {With the increasing ability of large language models (LLMs), in-context learning (ICL) has become a new paradigm for natural language processing (NLP), where LLMs make predictions only based on contexts augmented with a few examples. It has been a new trend to explore ICL to evaluate and extrapolate the ability of LLMs. In this paper, we aim to survey and summarize the progress and challenges of ICL. We first present a formal definition of ICL and clarify its correlation to related studies. Then, we organize and discuss advanced techniques, including training strategies, demonstration designing strategies, as well as related analysis. Finally, we discuss the challenges of ICL and provide potential directions for further research. We hope that our work can encourage more research on uncovering how ICL works and improving ICL.},\n\turldate = {2024-05-30},\n\tpublisher = {arXiv},\n\tauthor = {Dong, Qingxiu and Li, Lei and Dai, Damai and Zheng, Ce and Wu, Zhiyong and Chang, Baobao and Sun, Xu and Xu, Jingjing and Li, Lei and Sui, Zhifang},\n\tmonth = jun,\n\tyear = {2023},\n\tnote = {arXiv:2301.00234 [cs]},\n\tkeywords = {Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Jab/\\#Pre},\n}\n\n\n\n","author_short":["Dong, Q.","Li, L.","Dai, D.","Zheng, C.","Wu, Z.","Chang, B.","Sun, X.","Xu, J.","Li, L.","Sui, Z."],"key":"dong_survey_2023","id":"dong_survey_2023","bibbaseid":"dong-li-dai-zheng-wu-chang-sun-xu-etal-asurveyonincontextlearning-2023","role":"author","urls":{"Paper":"http://arxiv.org/abs/2301.00234"},"keyword":["Computer Science - Artificial Intelligence","Computer Science - Computation and Language","Jab/#Pre"],"metadata":{"authorlinks":{}},"downloads":0,"html":""},"bibtype":"misc","biburl":"https://bibbase.org/zotero/bxt101","dataSources":["Wsv2bQ4jPuc7qme8R"],"keywords":["computer science - artificial intelligence","computer science - computation and language","jab/#pre"],"search_terms":["survey","context","learning","dong","li","dai","zheng","wu","chang","sun","xu","li","sui"],"title":"A Survey on In-context Learning","year":2023}