{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:TGPGAX7WJSZNR7A4QYFKZC4BGY","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":"2c313efb5794d4d4071056045b3008be2ba12e5cab529e9401ed9b82c5957eb7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-31T16:55:06Z","title_canon_sha256":"8b0e5c1df72f88a5475a4ef8da559e4d21e83ea7b5bec580fa31cf433422efd1"},"schema_version":"1.0","source":{"id":"2310.20624","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.20624","created_at":"2026-07-05T08:21:42Z"},{"alias_kind":"arxiv_version","alias_value":"2310.20624v2","created_at":"2026-07-05T08:21:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.20624","created_at":"2026-07-05T08:21:42Z"},{"alias_kind":"pith_short_12","alias_value":"TGPGAX7WJSZN","created_at":"2026-07-05T08:21:42Z"},{"alias_kind":"pith_short_16","alias_value":"TGPGAX7WJSZNR7A4","created_at":"2026-07-05T08:21:42Z"},{"alias_kind":"pith_short_8","alias_value":"TGPGAX7W","created_at":"2026-07-05T08:21:42Z"}],"graph_snapshots":[{"event_id":"sha256:9da62099c9a438825a892bf72cac4ee87b2a45f68b9fe7a2ed5945d51e7fd558","target":"graph","created_at":"2026-07-05T08:21:42Z","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/2310.20624/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"AI developers often apply safety alignment procedures to prevent the misuse of their AI systems. For example, before Meta released Llama 2-Chat - a collection of instruction fine-tuned large language models - they invested heavily in safety training, incorporating extensive red-teaming and reinforcement learning from human feedback. We explore the robustness of safety training in language models by subversively fine-tuning Llama 2-Chat. We employ quantized low-rank adaptation (LoRA) as an efficient fine-tuning method. With a budget of less than \\$200 and using only one GPU, we successfully und","authors_text":"Charlie Rogers-Smith, Jeffrey Ladish, Simon Lermen","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-31T16:55:06Z","title":"LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.20624","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":"sha256:4ee1a1d8a8f85796b339252e902c442b1e52848810436dce61592f0eb76d5099","target":"record","created_at":"2026-07-05T08:21:42Z","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":"2c313efb5794d4d4071056045b3008be2ba12e5cab529e9401ed9b82c5957eb7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-31T16:55:06Z","title_canon_sha256":"8b0e5c1df72f88a5475a4ef8da559e4d21e83ea7b5bec580fa31cf433422efd1"},"schema_version":"1.0","source":{"id":"2310.20624","kind":"arxiv","version":2}},"canonical_sha256":"999e605ff64cb2d8fc1c860aac8b813639355d6b302a81e1c6b9e9a89a3eac76","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"999e605ff64cb2d8fc1c860aac8b813639355d6b302a81e1c6b9e9a89a3eac76","first_computed_at":"2026-07-05T08:21:42.728380Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:21:42.728380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CSYtiamAoFtt63Ws2hVsrLSBFmVOTeMIjU8ftgjOOykpPU04/uipAs3gaLr7NqqeMNn769NXdjhnnWwXy9okBg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:21:42.728877Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.20624","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4ee1a1d8a8f85796b339252e902c442b1e52848810436dce61592f0eb76d5099","sha256:9da62099c9a438825a892bf72cac4ee87b2a45f68b9fe7a2ed5945d51e7fd558"],"state_sha256":"6510d533ae4666d8623d1e25e55e52feba0d85b327615a64587a4158bcc63c17"}