{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:CPKJ2WVNVEIUDV7LWPFR3CREUE","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":"d9829cc3668d63018b3142f18355edda959f21154d21e97087b8d438f55dcb91","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-01T05:01:01Z","title_canon_sha256":"beed17607f96cecf05e6e0ec6ce2a2997b46a5e3220ad92402653b7ca76bd1f5"},"schema_version":"1.0","source":{"id":"2606.01695","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.01695","created_at":"2026-06-02T02:04:40Z"},{"alias_kind":"arxiv_version","alias_value":"2606.01695v1","created_at":"2026-06-02T02:04:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.01695","created_at":"2026-06-02T02:04:40Z"},{"alias_kind":"pith_short_12","alias_value":"CPKJ2WVNVEIU","created_at":"2026-06-02T02:04:40Z"},{"alias_kind":"pith_short_16","alias_value":"CPKJ2WVNVEIUDV7L","created_at":"2026-06-02T02:04:40Z"},{"alias_kind":"pith_short_8","alias_value":"CPKJ2WVN","created_at":"2026-06-02T02:04:40Z"}],"graph_snapshots":[{"event_id":"sha256:b0ac5d8f5c3b85baf2aad8d6e227739246afdc65e3fc6041e904ed6266252c10","target":"graph","created_at":"2026-06-02T02:04:40Z","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/2606.01695/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Adversaries can implant latent harmful behavior by poisoning as few as 1% of fine-tuning examples. The contamination is invisible to every output-level defense: harmful behavior lies dormant in the model's hidden-state geometry and does not appear in generated text until contamination exceeds 7.5%. We introduce CANARY (Contamination Auditor via Neural Activation Representation Yield), a zero-label checkpoint auditor that detects this hidden shift directly from two forward passes over an unlabeled prompt set. CANARY projects the hidden-state difference through a Sparse Autoencoder, filtering st","authors_text":"Swapnil Parekh","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-01T05:01:01Z","title":"CANARY: Zero-Label Detection of Fine-Tuning Contamination in Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.01695","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":"sha256:2c6216fcb484ecda85e515735cfd1af8d86a651e61481709cb42194f196280a7","target":"record","created_at":"2026-06-02T02:04:40Z","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":"d9829cc3668d63018b3142f18355edda959f21154d21e97087b8d438f55dcb91","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-01T05:01:01Z","title_canon_sha256":"beed17607f96cecf05e6e0ec6ce2a2997b46a5e3220ad92402653b7ca76bd1f5"},"schema_version":"1.0","source":{"id":"2606.01695","kind":"arxiv","version":1}},"canonical_sha256":"13d49d5aada91141d7ebb3cb1d8a24a120a1f935db842e99a97d007bdecf898d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"13d49d5aada91141d7ebb3cb1d8a24a120a1f935db842e99a97d007bdecf898d","first_computed_at":"2026-06-02T02:04:40.235451Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:40.235451Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sDdZ4gEQdJL7LSbwHMQHqg/Ux+OylOx1TDRvYFHNuGd8xnuxywoiatSJQN02394brp0r3DTxqibv1InEICyhBA==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:40.235900Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.01695","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2c6216fcb484ecda85e515735cfd1af8d86a651e61481709cb42194f196280a7","sha256:b0ac5d8f5c3b85baf2aad8d6e227739246afdc65e3fc6041e904ed6266252c10"],"state_sha256":"d377cfe94005cc4bdefc931d63307036051dcb99e10f1e105994cc6a2541f2a6"}