{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SSE4HMBJH37A3UWRKAJBLYZ4WG","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":"4f4f6487897f9f10b27c829909e46ccd33e8bd5dfdabc66981cea2fe142cc892","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-20T17:57:12Z","title_canon_sha256":"0a7772da67e976082a0a5708ee5c45c9cc174e515edbf39bd943f901c0dc7318"},"schema_version":"1.0","source":{"id":"2506.17209","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.17209","created_at":"2026-07-05T11:24:58Z"},{"alias_kind":"arxiv_version","alias_value":"2506.17209v1","created_at":"2026-07-05T11:24:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17209","created_at":"2026-07-05T11:24:58Z"},{"alias_kind":"pith_short_12","alias_value":"SSE4HMBJH37A","created_at":"2026-07-05T11:24:58Z"},{"alias_kind":"pith_short_16","alias_value":"SSE4HMBJH37A3UWR","created_at":"2026-07-05T11:24:58Z"},{"alias_kind":"pith_short_8","alias_value":"SSE4HMBJ","created_at":"2026-07-05T11:24:58Z"}],"graph_snapshots":[{"event_id":"sha256:1fe20211fd217299188d941502390fb8b34ae2f3c32111f8308731dbeb74a000","target":"graph","created_at":"2026-07-05T11:24:58Z","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/2506.17209/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning a general-purpose large language model (LLM) for a specific domain or task has become a routine procedure for ordinary users. However, fine-tuning is known to remove the safety alignment features of the model, even when the fine-tuning data does not contain any harmful content. We consider this to be a critical failure mode of LLMs due to the widespread uptake of fine-tuning, combined with the benign nature of the \"attack\". Most well-intentioned developers are likely unaware that they are deploying an LLM with reduced safety. On the other hand, this known vulnerability can be easil","authors_text":"Hillary Dawkins, Isar Nejadgholi, Kathleen C. Fraser, Svetlana Kiritchenko","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-20T17:57:12Z","title":"Fine-Tuning Lowers Safety and Disrupts Evaluation Consistency"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17209","kind":"arxiv","version":1},"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":"sha256:0268b7b17ca1dc506a2282c98dc6f2ebb2b6815aad30228cf3e38f9a03efe45b","target":"record","created_at":"2026-07-05T11:24:58Z","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":"4f4f6487897f9f10b27c829909e46ccd33e8bd5dfdabc66981cea2fe142cc892","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-20T17:57:12Z","title_canon_sha256":"0a7772da67e976082a0a5708ee5c45c9cc174e515edbf39bd943f901c0dc7318"},"schema_version":"1.0","source":{"id":"2506.17209","kind":"arxiv","version":1}},"canonical_sha256":"9489c3b0293efe0dd2d1501215e33cb18e5bccc90b1d80511fa1578f239e9984","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9489c3b0293efe0dd2d1501215e33cb18e5bccc90b1d80511fa1578f239e9984","first_computed_at":"2026-07-05T11:24:58.230305Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:24:58.230305Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eczfTRMm/ZRzvXC4BanqEp426JlR7T9DY3HyIhf03lzeYYE+DA60W0qK5WnoWskqsWzYl3ALnmi1nqQ/O8p5CA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:24:58.230846Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.17209","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0268b7b17ca1dc506a2282c98dc6f2ebb2b6815aad30228cf3e38f9a03efe45b","sha256:1fe20211fd217299188d941502390fb8b34ae2f3c32111f8308731dbeb74a000"],"state_sha256":"8b71ef5135648da198d4098123654ec45efa700c438042a71fda0e4c276b822e"}