{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:IFMVGDYBKIGOLAY267AZXD272A","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":"666154e42312cb28261b0614d512a9dcd15234a14cad9adc1a56d8fa3f1e754d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-24T18:44:28Z","title_canon_sha256":"dba388cbd399e0cdb7e237d0df45ea23c2730fcc11d23cbaf525cf9dd53d8fd4"},"schema_version":"1.0","source":{"id":"2507.18742","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.18742","created_at":"2026-07-05T11:42:57Z"},{"alias_kind":"arxiv_version","alias_value":"2507.18742v1","created_at":"2026-07-05T11:42:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.18742","created_at":"2026-07-05T11:42:57Z"},{"alias_kind":"pith_short_12","alias_value":"IFMVGDYBKIGO","created_at":"2026-07-05T11:42:57Z"},{"alias_kind":"pith_short_16","alias_value":"IFMVGDYBKIGOLAY2","created_at":"2026-07-05T11:42:57Z"},{"alias_kind":"pith_short_8","alias_value":"IFMVGDYB","created_at":"2026-07-05T11:42:57Z"}],"graph_snapshots":[{"event_id":"sha256:2d1b29b84392929fff78a82c73b9b365679cb049e262d456cf4e48ca2681eafe","target":"graph","created_at":"2026-07-05T11:42:57Z","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/2507.18742/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models (LMs) are susceptible to in-context reward hacking, where they exploit flaws in tainted or faulty written specifications or rubrics to achieve high scores without fulfilling the user's true intent. We introduce Specification Self-Correction (SSC), a novel, test-time framework that enables an LM to identify and correct flaws within its own guiding specification. SSC employs a multi-step inference process where the model first generates a response based on a potentially tainted specification, critiques its output, and then revises the specification itself to remove the exploitabl","authors_text":"V\\'ictor Gallego","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-24T18:44:28Z","title":"Specification Self-Correction: Mitigating In-Context Reward Hacking Through Test-Time Refinement"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.18742","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":"sha256:54ace307d54dc1b21fe6579cc599a8c54e226f14490abc8c15d9fd26db337427","target":"record","created_at":"2026-07-05T11:42:57Z","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":"666154e42312cb28261b0614d512a9dcd15234a14cad9adc1a56d8fa3f1e754d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-24T18:44:28Z","title_canon_sha256":"dba388cbd399e0cdb7e237d0df45ea23c2730fcc11d23cbaf525cf9dd53d8fd4"},"schema_version":"1.0","source":{"id":"2507.18742","kind":"arxiv","version":1}},"canonical_sha256":"4159530f01520ce5831af7c19b8f5fd01665cc8cd005abc1858b394803d029be","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4159530f01520ce5831af7c19b8f5fd01665cc8cd005abc1858b394803d029be","first_computed_at":"2026-07-05T11:42:57.132336Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:42:57.132336Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eQZPSJzXZDs7inPtXLuV7pfbg1KUVc0NPPZsAMxjOn+SXDDAbrciU54OqJ++JhuXTaHTpLmMJ6LiwuZQseu8DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:42:57.132794Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.18742","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:54ace307d54dc1b21fe6579cc599a8c54e226f14490abc8c15d9fd26db337427","sha256:2d1b29b84392929fff78a82c73b9b365679cb049e262d456cf4e48ca2681eafe"],"state_sha256":"60fb51a9958b2ee94a980562d96540a880f98e6b64d57da9fdb74cce56890530"}