{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:M3VKILJ7XNCXWHSSTM6C6N3OZB","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":"8cb5a280d25f85cd8b0cb4f6727edf1bd841c87af2254e60871d799eeab65538","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-07-11T17:52:03Z","title_canon_sha256":"e6145fcf1f807603cc82c041b521fcdfd7077f3b51e359656ef9ba4d34c6d262"},"schema_version":"1.0","source":{"id":"2407.08770","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.08770","created_at":"2026-07-05T10:12:17Z"},{"alias_kind":"arxiv_version","alias_value":"2407.08770v2","created_at":"2026-07-05T10:12:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08770","created_at":"2026-07-05T10:12:17Z"},{"alias_kind":"pith_short_12","alias_value":"M3VKILJ7XNCX","created_at":"2026-07-05T10:12:17Z"},{"alias_kind":"pith_short_16","alias_value":"M3VKILJ7XNCXWHSS","created_at":"2026-07-05T10:12:17Z"},{"alias_kind":"pith_short_8","alias_value":"M3VKILJ7","created_at":"2026-07-05T10:12:17Z"}],"graph_snapshots":[{"event_id":"sha256:c77f7b3314d3e848d587c1346ef6c84054819d0c4c7dcd5b2e1f074973984826","target":"graph","created_at":"2026-07-05T10:12:17Z","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/2407.08770/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have demonstrated great potential as generalist assistants, showcasing powerful task understanding and problem-solving capabilities. To deploy LLMs as AI assistants, it is crucial that these models exhibit desirable behavioral traits, such as non-toxicity and resilience against jailbreak attempts. Current approaches for detoxification or preventing jailbreaking usually involve Supervised Fine-Tuning (SFT) or Reinforcement Learning from Human Feedback (RLHF), which requires finetuning billions of parameters through gradient descent with substantial computational cos","authors_text":"Andrew Zhao, Gao Huang, Huanqian Wang, Jingxin Shi, Rui Lu, Shenzhi Wang, Shiji Song, Yang Yue","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-07-11T17:52:03Z","title":"Model Surgery: Modulating LLM's Behavior Via Simple Parameter Editing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08770","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":"sha256:51cf03b3ca69cac43452c7bfe6c9ae7b5e71cd5d2d0bb10c7300ab809ad2afa9","target":"record","created_at":"2026-07-05T10:12:17Z","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":"8cb5a280d25f85cd8b0cb4f6727edf1bd841c87af2254e60871d799eeab65538","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-07-11T17:52:03Z","title_canon_sha256":"e6145fcf1f807603cc82c041b521fcdfd7077f3b51e359656ef9ba4d34c6d262"},"schema_version":"1.0","source":{"id":"2407.08770","kind":"arxiv","version":2}},"canonical_sha256":"66eaa42d3fbb457b1e529b3c2f376ec867bad11d320e9d0710e14a04a76b613e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"66eaa42d3fbb457b1e529b3c2f376ec867bad11d320e9d0710e14a04a76b613e","first_computed_at":"2026-07-05T10:12:17.967389Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:12:17.967389Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zzIf/fTu28kxK/k8fv9HS6OPctX7QkvY9v0U4nmYmf2MCheGDuPwaNVPGPoUZHTZlm9a1sqOXdQLzkv0c1lBBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:12:17.967882Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.08770","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:51cf03b3ca69cac43452c7bfe6c9ae7b5e71cd5d2d0bb10c7300ab809ad2afa9","sha256:c77f7b3314d3e848d587c1346ef6c84054819d0c4c7dcd5b2e1f074973984826"],"state_sha256":"6501a1479d6849d06e6d112a85259d0c770bb70842db2ab2ad9228b1f4b67159"}