{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VPSA4TLXJDUIEFAG27IKSNKDOY","short_pith_number":"pith:VPSA4TLX","schema_version":"1.0","canonical_sha256":"abe40e4d7748e8821406d7d0a93543761dc9c0219a52612db4c9452c17f5df46","source":{"kind":"arxiv","id":"2607.01854","version":1},"attestation_state":"computed","paper":{"title":"Has This Checkpoint Been Abliterated? A Two-Signal Audit and Its Failure Map","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Gabriel Hurtado","submitted_at":"2026-07-02T08:15:25Z","abstract_excerpt":"Can a platform tell, before deployment, whether an open-weight checkpoint has had its refusal mechanism stripped? Runtime guards cannot: they score generations, not the artifact. We combine two cheap internal signals, a reference-anchored activation refusal-gap and a weight-recovery energy of the base-to-candidate weight difference, into a threshold-free checkpoint audit. The two are negatively correlated and label-complementary: the gap supplies refusal-specificity and the weight energy supplies recall. On a 273-checkpoint registry spanning Qwen, DeepSeek-distilled Qwen, Llama, and Gemma, the"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.01854","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-02T08:15:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"901c38b17917f997caa227392e8a34850f01dee9b5729a3d8663fbb947c97c35","abstract_canon_sha256":"8d837404ff50956bc5bb32b1392b482339d66a081706b208827d471e40b2dd36"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:31.720932Z","signature_b64":"JYhatbFAWCMifWtSeqvwM7JwjadFUz49Lndoh6VFRyinrM0xsGpghHQqo64lYQwzeFo/GVNH9qiF7Z3zT4YLAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abe40e4d7748e8821406d7d0a93543761dc9c0219a52612db4c9452c17f5df46","last_reissued_at":"2026-07-03T01:17:31.720564Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:31.720564Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Has This Checkpoint Been Abliterated? A Two-Signal Audit and Its Failure Map","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Gabriel Hurtado","submitted_at":"2026-07-02T08:15:25Z","abstract_excerpt":"Can a platform tell, before deployment, whether an open-weight checkpoint has had its refusal mechanism stripped? Runtime guards cannot: they score generations, not the artifact. We combine two cheap internal signals, a reference-anchored activation refusal-gap and a weight-recovery energy of the base-to-candidate weight difference, into a threshold-free checkpoint audit. The two are negatively correlated and label-complementary: the gap supplies refusal-specificity and the weight energy supplies recall. On a 273-checkpoint registry spanning Qwen, DeepSeek-distilled Qwen, Llama, and Gemma, the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.01854","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.01854/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.01854","created_at":"2026-07-03T01:17:31.720630+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.01854v1","created_at":"2026-07-03T01:17:31.720630+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.01854","created_at":"2026-07-03T01:17:31.720630+00:00"},{"alias_kind":"pith_short_12","alias_value":"VPSA4TLXJDUI","created_at":"2026-07-03T01:17:31.720630+00:00"},{"alias_kind":"pith_short_16","alias_value":"VPSA4TLXJDUIEFAG","created_at":"2026-07-03T01:17:31.720630+00:00"},{"alias_kind":"pith_short_8","alias_value":"VPSA4TLX","created_at":"2026-07-03T01:17:31.720630+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY","json":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY.json","graph_json":"https://pith.science/api/pith-number/VPSA4TLXJDUIEFAG27IKSNKDOY/graph.json","events_json":"https://pith.science/api/pith-number/VPSA4TLXJDUIEFAG27IKSNKDOY/events.json","paper":"https://pith.science/paper/VPSA4TLX"},"agent_actions":{"view_html":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY","download_json":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY.json","view_paper":"https://pith.science/paper/VPSA4TLX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.01854&json=true","fetch_graph":"https://pith.science/api/pith-number/VPSA4TLXJDUIEFAG27IKSNKDOY/graph.json","fetch_events":"https://pith.science/api/pith-number/VPSA4TLXJDUIEFAG27IKSNKDOY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY/action/storage_attestation","attest_author":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY/action/author_attestation","sign_citation":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY/action/citation_signature","submit_replication":"https://pith.science/pith/VPSA4TLXJDUIEFAG27IKSNKDOY/action/replication_record"}},"created_at":"2026-07-03T01:17:31.720630+00:00","updated_at":"2026-07-03T01:17:31.720630+00:00"}