{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AY2RYV7V75NGA5QUTPZFGQLQQ4","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":"1620b8ceaf4e5ccec2e117b7d5c93f0c39c3067ee927d9d4395ee2dc6511c1b5","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-07-24T17:59:24Z","title_canon_sha256":"d8083ce27bc1bc58c0eb97399a408512142d74b1ba1837b1fccb201e4e435d77"},"schema_version":"1.0","source":{"id":"2507.18631","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.18631","created_at":"2026-07-05T11:43:15Z"},{"alias_kind":"arxiv_version","alias_value":"2507.18631v2","created_at":"2026-07-05T11:43:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.18631","created_at":"2026-07-05T11:43:15Z"},{"alias_kind":"pith_short_12","alias_value":"AY2RYV7V75NG","created_at":"2026-07-05T11:43:15Z"},{"alias_kind":"pith_short_16","alias_value":"AY2RYV7V75NGA5QU","created_at":"2026-07-05T11:43:15Z"},{"alias_kind":"pith_short_8","alias_value":"AY2RYV7V","created_at":"2026-07-05T11:43:15Z"}],"graph_snapshots":[{"event_id":"sha256:2e6c9822e5a516212f6030aa67afb002cd34c6b6999c74951234596a855090ca","target":"graph","created_at":"2026-07-05T11:43:15Z","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/2507.18631/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With rapid advancement and increasing accessibility of LLMs, fine-tuning aligned models has become a critical step for adapting them to real-world applications, which makes the safety of this fine-tuning process more important than ever. However, recent studies have highlighted a critical challenge: even when fine-tuning with seemingly benign downstream datasets, the safety of aligned LLMs can be compromised, making them more susceptible to malicious instructions.\n  In this paper, we show that fine-tuning datasets often contain samples with safety-degrading features that are not easily identif","authors_text":"Hao Li, Jing Shao, Lei Sha, Lijun Li, Rui Li, Xianyi Wei, Zhenghao Lu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-07-24T17:59:24Z","title":"Layer-Aware Representation Filtering: Purifying Finetuning Data to Preserve LLM Safety Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.18631","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":"sha256:3eb590ecd0d0fb36027bf6f99522ed66b9646717bfc17685e717f98d9fbc446c","target":"record","created_at":"2026-07-05T11:43:15Z","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":"1620b8ceaf4e5ccec2e117b7d5c93f0c39c3067ee927d9d4395ee2dc6511c1b5","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-07-24T17:59:24Z","title_canon_sha256":"d8083ce27bc1bc58c0eb97399a408512142d74b1ba1837b1fccb201e4e435d77"},"schema_version":"1.0","source":{"id":"2507.18631","kind":"arxiv","version":2}},"canonical_sha256":"06351c57f5ff5a6076149bf2534170872e9c819859ec37f35302c6976886e7cc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"06351c57f5ff5a6076149bf2534170872e9c819859ec37f35302c6976886e7cc","first_computed_at":"2026-07-05T11:43:15.041655Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:43:15.041655Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7XA+nPE+asv0FFlGn0AvKcPaYgsCP4ZvXR/jr3/aPu2qeiWUkinbDc9Z+xylEBtfu08qxIQ1v5eNLaJ65zyhAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:43:15.042120Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.18631","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3eb590ecd0d0fb36027bf6f99522ed66b9646717bfc17685e717f98d9fbc446c","sha256:2e6c9822e5a516212f6030aa67afb002cd34c6b6999c74951234596a855090ca"],"state_sha256":"6b7ed7488af271b693517d2d51e438ed9d340dc035a4a9b7df3f3082739785a4"}