{"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/adventure-adversarial-training-for-textual","title":"AdvEntuRe: Adversarial Training for Textual Entailment with Knowledge-Guided Examples","arxiv_id":"1805.04680","date":"2018-05-12","proceeding":"ACL 2018 7","authors":["Dongyeop Kang","Tushar Khot","Ashish Sabharwal","Eduard Hovy"],"abstract":"We consider the problem of learning textual entailment models with limited\nsupervision (5K-10K training examples), and present two complementary\napproaches for it. First, we propose knowledge-guided adversarial example\ngenerators for incorporating large lexical resources in entailment models via\nonly a handful of rule templates. Second, to make the entailment model - a\ndiscriminator - more robust, we propose the first GAN-style approach for\ntraining it using a natural language example generator that iteratively adjusts\nbased on the discriminator's performance. We demonstrate effectiveness using\ntwo entailment datasets, where the proposed methods increase accuracy by 4.7%\non SciTail and by 2.8% on a 1% training sub-sample of SNLI. Notably, even a\nsingle hand-written rule, negate, improves the accuracy on the negation\nexamples in SNLI by 6.1%.","url_abs":"http://arxiv.org/abs/1805.04680v1","url_pdf":"http://arxiv.org/pdf/1805.04680v1.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":"adventure-adversarial-training-for-textual","repo_url":"https://github.com/dykang/adventure","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"task_slug":"negation","task_name":"Negation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.04680","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}