{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WNLHMCPU3H4JIZ6DE27B57DOUE","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":"74ee4a4804ef4e80206410d1dc786e83d4788f32a4962c8808e09c840da04a77","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-29T20:40:21Z","title_canon_sha256":"b88f8edd953240022685755778c3610b15f7179f3281a43eeb90442cfbb02d97"},"schema_version":"1.0","source":{"id":"2606.30899","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.30899","created_at":"2026-07-01T00:17:20Z"},{"alias_kind":"arxiv_version","alias_value":"2606.30899v1","created_at":"2026-07-01T00:17:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.30899","created_at":"2026-07-01T00:17:20Z"},{"alias_kind":"pith_short_12","alias_value":"WNLHMCPU3H4J","created_at":"2026-07-01T00:17:20Z"},{"alias_kind":"pith_short_16","alias_value":"WNLHMCPU3H4JIZ6D","created_at":"2026-07-01T00:17:20Z"},{"alias_kind":"pith_short_8","alias_value":"WNLHMCPU","created_at":"2026-07-01T00:17:20Z"}],"graph_snapshots":[{"event_id":"sha256:0e9c1d2f3197c086d48766afe60bdad1a6dc8a0692b650e6b67091f1927b4acf","target":"graph","created_at":"2026-07-01T00:17:20Z","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/2606.30899/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Backdoor attacks pose a serious threat to large language models (LLMs) by causing otherwise benign systems to produce attacker-specified malicious behavior when a hidden trigger is present. In this work, we study post hoc detoxification of backdoored LLMs in a practical setting where the defender has access to the poisoned model but does not wish to retrain the full network from scratch. We propose a mechanistically guided weight-space repair framework that first localizes modules involved in propagating trigger-induced behavior using activation patching and Fisher/K-FAC curvature analysis, an","authors_text":"Andrew Arash Mahyari, Arash Raftari, Mehrdad Mahdavi, Nathan Blackthorn","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-29T20:40:21Z","title":"Curvature-Guided Module Localization for Low-Rank Detoxification of Backdoored Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.30899","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":"sha256:1e34008e702de019a48376b99ba77197defd6d373313fb7166980f98238d6e54","target":"record","created_at":"2026-07-01T00:17:20Z","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":"74ee4a4804ef4e80206410d1dc786e83d4788f32a4962c8808e09c840da04a77","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-29T20:40:21Z","title_canon_sha256":"b88f8edd953240022685755778c3610b15f7179f3281a43eeb90442cfbb02d97"},"schema_version":"1.0","source":{"id":"2606.30899","kind":"arxiv","version":1}},"canonical_sha256":"b3567609f4d9f89467c326be1efc6ea13a8cd516d9c9ecef882362630ff25532","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b3567609f4d9f89467c326be1efc6ea13a8cd516d9c9ecef882362630ff25532","first_computed_at":"2026-07-01T00:17:20.872445Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-01T00:17:20.872445Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9aVKLPWhPBq0yXHIotkqnAQev+1/1mdKuaOfanOPglTvR3RzcUw9as3ALtPNKvGQ7uF4+DNxQgCrIqsRkTMABQ==","signature_status":"signed_v1","signed_at":"2026-07-01T00:17:20.872902Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.30899","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1e34008e702de019a48376b99ba77197defd6d373313fb7166980f98238d6e54","sha256:0e9c1d2f3197c086d48766afe60bdad1a6dc8a0692b650e6b67091f1927b4acf"],"state_sha256":"a901487bde6af8ebb238215b669421109f8f48a96cf90a4516cfe3d5761c905e"}