{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/e-snli-natural-language-inference-with","title":"e-SNLI: Natural Language Inference with Natural Language Explanations","arxiv_id":"1812.01193","date":"2018-12-04","proceeding":"NeurIPS 2018 12","authors":["Oana-Maria Camburu","Tim Rocktäschel","Thomas Lukasiewicz","Phil Blunsom"],"abstract":"In order for machine learning to garner widespread public adoption, models\nmust be able to provide interpretable and robust explanations for their\ndecisions, as well as learn from human-provided explanations at train time. In\nthis work, we extend the Stanford Natural Language Inference dataset with an\nadditional layer of human-annotated natural language explanations of the\nentailment relations. We further implement models that incorporate these\nexplanations into their training process and output them at test time. We show\nhow our corpus of explanations, which we call e-SNLI, can be used for various\ngoals, such as obtaining full sentence justifications of a model's decisions,\nimproving universal sentence representations and transferring to out-of-domain\nNLI datasets. Our dataset thus opens up a range of research directions for\nusing natural language explanations, both for improving models and for\nasserting their trust.","url_abs":"http://arxiv.org/abs/1812.01193v2","url_pdf":"http://arxiv.org/pdf/1812.01193v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"e-snli-natural-language-inference-with","repo_url":"https://github.com/OanaMariaCamburu/e-SNLI","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"e-snli-natural-language-inference-with","repo_url":"https://github.com/qtli/eib","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[{"slug":"e-snli","name":"e-SNLI","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/natural-language-inference-on-e-snli","task":"Natural Language Inference","dataset":"e-SNLI","model":"ExplainThenPredictAttention (e-InferSent Bi-LSTM + Attention)","rank_in_archive_order":1,"of":3,"metrics":{"Accuracy":"81.71","BLEU":"27.58"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1812.01193","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}