{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VECC3CTY75XFNNOS2RBKCO7DNS","short_pith_number":"pith:VECC3CTY","schema_version":"1.0","canonical_sha256":"a9042d8a78ff6e56b5d2d442a13be36cae3d87a934a4e28c48fe6587d1a225b3","source":{"kind":"arxiv","id":"2606.20740","version":1},"attestation_state":"computed","paper":{"title":"VeriBound: PAC-Bayesian Generalization Bounds for Process Reward Models Trained with Formal Verification Tools","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Amirul Rahman, Mohammed Sabih Alsharari","submitted_at":"2026-06-17T19:05:38Z","abstract_excerpt":"Process Reward Models (PRMs) provide step-level verification for Large Language Model (LLM) reasoning, yet their training data acquisition remains a bottleneck: human annotation is costly and Monte Carlo roll-out estimates are noisy. A recent approach, FOVER, trains PRMs on step-level error labels automatically annotated by formal verification tools such as Z3 and Isabelle, and empirically observes cross-task generalization from symbolic tasks to diverse reasoning benchmarks. However, this generalization phenomenon lacks any theoretical explanation, and no formal bounds exist on the generaliza"},"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":"2606.20740","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-17T19:05:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7d233a7b7a6ad2d663393bf44f5fc1ec29d1797b69ba4329e939955292584189","abstract_canon_sha256":"c41f3886a3abedd20b4fba3b8883ca01f2b84cc60bdbaefa90d4b2970fcd7998"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:55.236663Z","signature_b64":"JlhWFmw5ssK2zqnrrtIl8IzRnx1o9OvZUpHal9K3yNPVjDVXlxAZM89qZs8s0VgF5Alj3wg/ShfvS6NvjgnNAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a9042d8a78ff6e56b5d2d442a13be36cae3d87a934a4e28c48fe6587d1a225b3","last_reissued_at":"2026-06-23T00:11:55.236284Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:55.236284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VeriBound: PAC-Bayesian Generalization Bounds for Process Reward Models Trained with Formal Verification Tools","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Amirul Rahman, Mohammed Sabih Alsharari","submitted_at":"2026-06-17T19:05:38Z","abstract_excerpt":"Process Reward Models (PRMs) provide step-level verification for Large Language Model (LLM) reasoning, yet their training data acquisition remains a bottleneck: human annotation is costly and Monte Carlo roll-out estimates are noisy. A recent approach, FOVER, trains PRMs on step-level error labels automatically annotated by formal verification tools such as Z3 and Isabelle, and empirically observes cross-task generalization from symbolic tasks to diverse reasoning benchmarks. However, this generalization phenomenon lacks any theoretical explanation, and no formal bounds exist on the generaliza"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20740","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/2606.20740/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":"2606.20740","created_at":"2026-06-23T00:11:55.236349+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20740v1","created_at":"2026-06-23T00:11:55.236349+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20740","created_at":"2026-06-23T00:11:55.236349+00:00"},{"alias_kind":"pith_short_12","alias_value":"VECC3CTY75XF","created_at":"2026-06-23T00:11:55.236349+00:00"},{"alias_kind":"pith_short_16","alias_value":"VECC3CTY75XFNNOS","created_at":"2026-06-23T00:11:55.236349+00:00"},{"alias_kind":"pith_short_8","alias_value":"VECC3CTY","created_at":"2026-06-23T00:11:55.236349+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/VECC3CTY75XFNNOS2RBKCO7DNS","json":"https://pith.science/pith/VECC3CTY75XFNNOS2RBKCO7DNS.json","graph_json":"https://pith.science/api/pith-number/VECC3CTY75XFNNOS2RBKCO7DNS/graph.json","events_json":"https://pith.science/api/pith-number/VECC3CTY75XFNNOS2RBKCO7DNS/events.json","paper":"https://pith.science/paper/VECC3CTY"},"agent_actions":{"view_html":"https://pith.science/pith/VECC3CTY75XFNNOS2RBKCO7DNS","download_json":"https://pith.science/pith/VECC3CTY75XFNNOS2RBKCO7DNS.json","view_paper":"https://pith.science/paper/VECC3CTY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20740&json=true","fetch_graph":"https://pith.science/api/pith-number/VECC3CTY75XFNNOS2RBKCO7DNS/graph.json","fetch_events":"https://pith.science/api/pith-number/VECC3CTY75XFNNOS2RBKCO7DNS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VECC3CTY75XFNNOS2RBKCO7DNS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VECC3CTY75XFNNOS2RBKCO7DNS/action/storage_attestation","attest_author":"https://pith.science/pith/VECC3CTY75XFNNOS2RBKCO7DNS/action/author_attestation","sign_citation":"https://pith.science/pith/VECC3CTY75XFNNOS2RBKCO7DNS/action/citation_signature","submit_replication":"https://pith.science/pith/VECC3CTY75XFNNOS2RBKCO7DNS/action/replication_record"}},"created_at":"2026-06-23T00:11:55.236349+00:00","updated_at":"2026-06-23T00:11:55.236349+00:00"}