{"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/jailbreaking-is-best-solved-by-definition","title":"Testing the Limits of Jailbreaking Defenses with the Purple Problem","arxiv_id":"2403.14725","date":"2024-03-20","proceeding":null,"authors":["Taeyoun Kim","Suhas Kotha","aditi raghunathan"],"abstract":"The rise of \"jailbreak\" attacks on language models has led to a flurry of defenses aimed at preventing undesirable responses. We critically examine the two stages of the defense pipeline: (i) defining what constitutes unsafe outputs, and (ii) enforcing the definition via methods such as input processing or fine-tuning. To test the efficacy of existing enforcement mechanisms, we consider a simple and well-specified definition of unsafe outputs--outputs that contain the word \"purple\". Surprisingly, existing fine-tuning and input defenses fail on this simple problem, casting doubt on whether enforcement algorithms can be robust for more complicated definitions. We find that real safety benchmarks similarly test enforcement for a fixed definition. We hope that future research can lead to effective/fast enforcement as well as high quality definitions used for enforcement and evaluation.","url_abs":"https://arxiv.org/abs/2403.14725v2","url_pdf":"https://arxiv.org/pdf/2403.14725v2.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":"jailbreaking-is-best-solved-by-definition","repo_url":"https://github.com/kothasuhas/purple-problem","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2403.14725","atlas_url":"https://app.syntology.ai/?focus=2403.14725","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.14725"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kothasuhas/purple-problem","reach":null}],"summary":{"ran_draft_wrong":2,"unverified":1},"by_repo_kind":{"official":{"samples":3,"ran":2,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"925e98db3714f330","entry":"get_adversarial_suffix","repo":"kothasuhas/purple-problem","repo_kind":"official","path":"train_dpo.py","file_url":"https://github.com/kothasuhas/purple-problem/blob/HEAD/train_dpo.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"925e98db3714f330"}},{"code_sha256_prefix":"5dc945d2536d6656","entry":"load_dataset","repo":"kothasuhas/purple-problem","repo_kind":"official","path":"evaluate.py","file_url":"https://github.com/kothasuhas/purple-problem/blob/HEAD/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5dc945d2536d6656"}},{"code_sha256_prefix":"1730e540ac46b681","entry":"get_log_perplexity","repo":"kothasuhas/purple-problem","repo_kind":"official","path":"evaluate.py","file_url":"https://github.com/kothasuhas/purple-problem/blob/HEAD/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1730e540ac46b681"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}