{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:5INDX4EMZI4QUZP4M4K2SIYLFP","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":"51e7174afc18505f95a083baea756739a282d8feddf28468653789d2c305ccbc","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-16T17:02:53Z","title_canon_sha256":"de32337f9a4350c3dd664c69a6d70f0ec94b007239227541958aab2f66c1b772"},"schema_version":"1.0","source":{"id":"2403.04769","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.04769","created_at":"2026-07-05T07:54:09Z"},{"alias_kind":"arxiv_version","alias_value":"2403.04769v2","created_at":"2026-07-05T07:54:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.04769","created_at":"2026-07-05T07:54:09Z"},{"alias_kind":"pith_short_12","alias_value":"5INDX4EMZI4Q","created_at":"2026-07-05T07:54:09Z"},{"alias_kind":"pith_short_16","alias_value":"5INDX4EMZI4QUZP4","created_at":"2026-07-05T07:54:09Z"},{"alias_kind":"pith_short_8","alias_value":"5INDX4EM","created_at":"2026-07-05T07:54:09Z"}],"graph_snapshots":[{"event_id":"sha256:ab8f9580ddea01e629d7573753f79b6e4468e145ca8237a69a99fb8e65c3f0a3","target":"graph","created_at":"2026-07-05T07:54:09Z","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/2403.04769/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are initially trained on vast amounts of data, then fine-tuned using reinforcement learning from human feedback (RLHF); this also serves to teach the LLM to provide appropriate and safe responses. In this paper, we present a novel method to manipulate the fine-tuned version into reverting to its pre-RLHF behavior, effectively erasing the model's filters; the exploit currently works for GPT4, Claude Sonnet, and (to some extent) for Inflection-2.5. Unlike other jailbreaks (for example, the popular \"Do Anything Now\" (DAN) ), our method does not rely on instructing the","authors_text":"Benjamin Lemkin","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-16T17:02:53Z","title":"Using Hallucinations to Bypass GPT4's Filter"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.04769","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":"sha256:2ab171effceb325c577a3f71524979c56eec6d6696af855f12228ff80a346301","target":"record","created_at":"2026-07-05T07:54:09Z","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":"51e7174afc18505f95a083baea756739a282d8feddf28468653789d2c305ccbc","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-16T17:02:53Z","title_canon_sha256":"de32337f9a4350c3dd664c69a6d70f0ec94b007239227541958aab2f66c1b772"},"schema_version":"1.0","source":{"id":"2403.04769","kind":"arxiv","version":2}},"canonical_sha256":"ea1a3bf08cca390a65fc6715a9230b2bcb4a770960c69caa0b3e641569457815","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ea1a3bf08cca390a65fc6715a9230b2bcb4a770960c69caa0b3e641569457815","first_computed_at":"2026-07-05T07:54:09.546805Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:54:09.546805Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1AmUkbnz0yrouYVexWtpclQ5vKIksRAULe/MMNm9X4Hg4vSxO3MYpAf4uwqft0DODLGeihuOQI9+YVxMjMt2Cg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:54:09.547372Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.04769","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2ab171effceb325c577a3f71524979c56eec6d6696af855f12228ff80a346301","sha256:ab8f9580ddea01e629d7573753f79b6e4468e145ca8237a69a99fb8e65c3f0a3"],"state_sha256":"e50d9820a8e314d0521253931614f5d20d5629992b1063e04adb841bde75ed39"}