{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZAWQFZTLG37SLOERYKFIYUOTE5","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":"457ce90e5d7a30859c98387255c0a0ee463280e163eed8341178b0f4db5c3654","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-10T04:13:41Z","title_canon_sha256":"38763320507914b38311c43474779d790328c24b137bc60863cab7ca463a47d0"},"schema_version":"1.0","source":{"id":"2508.07172","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.07172","created_at":"2026-07-05T11:51:41Z"},{"alias_kind":"arxiv_version","alias_value":"2508.07172v1","created_at":"2026-07-05T11:51:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07172","created_at":"2026-07-05T11:51:41Z"},{"alias_kind":"pith_short_12","alias_value":"ZAWQFZTLG37S","created_at":"2026-07-05T11:51:41Z"},{"alias_kind":"pith_short_16","alias_value":"ZAWQFZTLG37SLOER","created_at":"2026-07-05T11:51:41Z"},{"alias_kind":"pith_short_8","alias_value":"ZAWQFZTL","created_at":"2026-07-05T11:51:41Z"}],"graph_snapshots":[{"event_id":"sha256:eef1d6b3a7428e0d6b3252679d801a9a3e199bc3aa8776121dbf1aff9d1f6f7f","target":"graph","created_at":"2026-07-05T11:51:41Z","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/2508.07172/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning-as-a-Service introduces a critical vulnerability where a few malicious examples mixed into the user's fine-tuning dataset can compromise the safety alignment of Large Language Models (LLMs). While a recognized paradigm frames safe fine-tuning as a multi-objective optimization problem balancing user task performance with safety alignment, we find existing solutions are critically sensitive to the harmful ratio, with defenses degrading sharply as harmful ratio increases. We diagnose that this failure stems from conflicting gradients, where the user-task update directly undermines the","authors_text":"Baolei Zhang, Biao Yi, Jiahao Li, Lihai Nie, Tiansheng Huang, Tong Li, Zheli Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-10T04:13:41Z","title":"Gradient Surgery for Safe LLM Fine-Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07172","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":"sha256:6ea5743c6c78d7a8b387f92967d29a22f04674af19a04364683f863dfbcb6eee","target":"record","created_at":"2026-07-05T11:51:41Z","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":"457ce90e5d7a30859c98387255c0a0ee463280e163eed8341178b0f4db5c3654","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-10T04:13:41Z","title_canon_sha256":"38763320507914b38311c43474779d790328c24b137bc60863cab7ca463a47d0"},"schema_version":"1.0","source":{"id":"2508.07172","kind":"arxiv","version":1}},"canonical_sha256":"c82d02e66b36ff25b891c28a8c51d3274893b02cdfdbb0958669b842b5e4e3cd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c82d02e66b36ff25b891c28a8c51d3274893b02cdfdbb0958669b842b5e4e3cd","first_computed_at":"2026-07-05T11:51:41.955966Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:51:41.955966Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"m9ecivyqFIOiI8YxTkPhC84NrE/DMAnpi9LiQZxBHDkLyjJoqU4krKNcNWssVOJIu21Ly006IIDL6Alr8IHWBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:51:41.956441Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.07172","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6ea5743c6c78d7a8b387f92967d29a22f04674af19a04364683f863dfbcb6eee","sha256:eef1d6b3a7428e0d6b3252679d801a9a3e199bc3aa8776121dbf1aff9d1f6f7f"],"state_sha256":"e5ee1b30f6ef1d3e07451564fd81b99d683d20645e66e00c049786af31a03485"}