{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:EGLX34FCUMWYMWTDRWTS3ITFVZ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e2e70b0eecd10ebb7780a99c4453d414577ed65f4649cdb2665123ae57aae437","cross_cats_sorted":["cs.CR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-28T19:16:17Z","title_canon_sha256":"593e5a7335985fae38678c90fc3c09982246292c010560667ec1d7b6f7c798cc"},"schema_version":"1.0","source":{"id":"2405.18540","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.18540","created_at":"2026-07-05T10:21:19Z"},{"alias_kind":"arxiv_version","alias_value":"2405.18540v2","created_at":"2026-07-05T10:21:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.18540","created_at":"2026-07-05T10:21:19Z"},{"alias_kind":"pith_short_12","alias_value":"EGLX34FCUMWY","created_at":"2026-07-05T10:21:19Z"},{"alias_kind":"pith_short_16","alias_value":"EGLX34FCUMWYMWTD","created_at":"2026-07-05T10:21:19Z"},{"alias_kind":"pith_short_8","alias_value":"EGLX34FC","created_at":"2026-07-05T10:21:19Z"}],"graph_snapshots":[{"event_id":"sha256:1cf1f27fcee2364862eae7f27a94a1ff074e6d06f81e33250369bd946cbdb2b7","target":"graph","created_at":"2026-07-05T10:21:19Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2405.18540/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Red-teaming, or identifying prompts that elicit harmful responses, is a critical step in ensuring the safe and responsible deployment of large language models (LLMs). Developing effective protection against many modes of attack prompts requires discovering diverse attacks. Automated red-teaming typically uses reinforcement learning to fine-tune an attacker language model to generate prompts that elicit undesirable responses from a target LLM, as measured, for example, by an auxiliary toxicity classifier. We show that even with explicit regularization to favor novelty and diversity, existing ap","authors_text":"David Dobre, Gauthier Gidel, Juho Lee, Kenji Kawaguchi, Lynn Cherif, Minsu Kim, Moksh Jain, Nikolay Malkin, Seanie Lee, Sung Ju Hwang, Yoshua Bengio","cross_cats":["cs.CR","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-28T19:16:17Z","title":"Learning diverse attacks on large language models for robust red-teaming and safety tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.18540","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a8062400c72802dd3b40e27967be35c05fec86aa50cbf971e993893c94412300","target":"record","created_at":"2026-07-05T10:21:19Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e2e70b0eecd10ebb7780a99c4453d414577ed65f4649cdb2665123ae57aae437","cross_cats_sorted":["cs.CR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-28T19:16:17Z","title_canon_sha256":"593e5a7335985fae38678c90fc3c09982246292c010560667ec1d7b6f7c798cc"},"schema_version":"1.0","source":{"id":"2405.18540","kind":"arxiv","version":2}},"canonical_sha256":"21977df0a2a32d865a638da72da265ae72d3f23e2897f4771abb242142571b14","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"21977df0a2a32d865a638da72da265ae72d3f23e2897f4771abb242142571b14","first_computed_at":"2026-07-05T10:21:19.568097Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:21:19.568097Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DJHXsyMYqY8QcQsrAibwa4eR0s1Y/9H82v4OMEGFpRp7qhO/kGRNg/KbbaxaMzwew/7ldvmxkOQROQj5DREMAA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:21:19.568620Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.18540","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a8062400c72802dd3b40e27967be35c05fec86aa50cbf971e993893c94412300","sha256:1cf1f27fcee2364862eae7f27a94a1ff074e6d06f81e33250369bd946cbdb2b7"],"state_sha256":"42e55f43474f10c6fbd417dce944f34a5fda6ce911fd643f072937f9490ae66f"}