{"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/dancin-seq2seq-fooling-text-classifiers-with","title":"DANCin SEQ2SEQ: Fooling Text Classifiers with Adversarial Text Example Generation","arxiv_id":"1712.05419","date":"2017-12-14","proceeding":null,"authors":["Catherine Wong"],"abstract":"Machine learning models are powerful but fallible. Generating adversarial\nexamples - inputs deliberately crafted to cause model misclassification or\nother errors - can yield important insight into model assumptions and\nvulnerabilities. Despite significant recent work on adversarial example\ngeneration targeting image classifiers, relatively little work exists exploring\nadversarial example generation for text classifiers; additionally, many\nexisting adversarial example generation algorithms require full access to\ntarget model parameters, rendering them impractical for many real-world\nattacks. In this work, we introduce DANCin SEQ2SEQ, a GAN-inspired algorithm\nfor adversarial text example generation targeting largely black-box text\nclassifiers. We recast adversarial text example generation as a reinforcement\nlearning problem, and demonstrate that our algorithm offers preliminary but\npromising steps towards generating semantically meaningful adversarial text\nexamples in a real-world attack scenario.","url_abs":"http://arxiv.org/abs/1712.05419v1","url_pdf":"http://arxiv.org/pdf/1712.05419v1.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":"dancin-seq2seq-fooling-text-classifiers-with","repo_url":"https://github.com/CatherineWong/dancin_seq2seq","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"adversarial-text","task_name":"Adversarial Text"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.05419","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}