{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DJBI4SCTW4RQNRTYKREECMCZOZ","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":"9d1f0e7d16fb70f33a8145e6c5a1e164d0983e0650cc28ee0baa03cd8e2e6230","cross_cats_sorted":["cs.AI","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-27T06:36:22Z","title_canon_sha256":"9c206e5fd9de4446ad7ff16ffec85fc31513e7c826326b15835e218b82b0e34a"},"schema_version":"1.0","source":{"id":"2605.28030","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28030","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28030v1","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28030","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_12","alias_value":"DJBI4SCTW4RQ","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_16","alias_value":"DJBI4SCTW4RQNRTY","created_at":"2026-05-28T01:04:56Z"},{"alias_kind":"pith_short_8","alias_value":"DJBI4SCT","created_at":"2026-05-28T01:04:56Z"}],"graph_snapshots":[{"event_id":"sha256:1974d80af800e873c8cbe2e9321315e12951a70c23f14cd42933a5aa9e8a189d","target":"graph","created_at":"2026-05-28T01:04:56Z","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/2605.28030/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning large language models often undermines their safety alignment, a problem further amplified by harmful fine-tuning attacks in which adversarial data removes safeguards and induces unsafe behaviors. We propose SPARD, a defense framework that integrates Safety-Projected Alternating optimization with Relevance-Diversity aware data selection. SPARD employs SPAG, which optimizes alternatively between utility updates and explicit safety projections with a set of safe data to enforce safety constraints. To curate safe data, we introduce a Relevance-Diversity Determinantal Point Process to ","authors_text":"Chengxiang Zhuo, James T. Kwok, Shengda Luo, Shuhao Chen, Weisen Jiang, Yeqi Gong, Yu Zhang, Zang Li","cross_cats":["cs.AI","cs.CR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-27T06:36:22Z","title":"SPARD: Defending Harmful Fine-Tuning Attack via Safety Projection with Relevance-Diversity Data Selection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28030","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":"sha256:2fcb0436ff69f33dafa3b16552c74642a4dac5b890bb3ba748815f82e9dfc39b","target":"record","created_at":"2026-05-28T01:04:56Z","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":"9d1f0e7d16fb70f33a8145e6c5a1e164d0983e0650cc28ee0baa03cd8e2e6230","cross_cats_sorted":["cs.AI","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-27T06:36:22Z","title_canon_sha256":"9c206e5fd9de4446ad7ff16ffec85fc31513e7c826326b15835e218b82b0e34a"},"schema_version":"1.0","source":{"id":"2605.28030","kind":"arxiv","version":1}},"canonical_sha256":"1a428e4853b72306c6785448413059767edeac296d6a50e761fa76b53e969c90","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1a428e4853b72306c6785448413059767edeac296d6a50e761fa76b53e969c90","first_computed_at":"2026-05-28T01:04:56.482226Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:56.482226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"b+DmOIxY9XhRT1krhSpOKNWqpCKzU3G0Ebg5ZpsrSOOVZkEmeWSI62fyv+yUKIn2WL0SBzzXkFkxWb8Qr2VRAg==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:56.482625Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.28030","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2fcb0436ff69f33dafa3b16552c74642a4dac5b890bb3ba748815f82e9dfc39b","sha256:1974d80af800e873c8cbe2e9321315e12951a70c23f14cd42933a5aa9e8a189d"],"state_sha256":"fc27f7031ffff622f08ef3801de29afcfff9fa642e5ace87bda9fdbab8fc8505"}