{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:P22VHLU7DDCJG4XGBEEOCE5VQT","short_pith_number":"pith:P22VHLU7","schema_version":"1.0","canonical_sha256":"7eb553ae9f18c49372e60908e113b584fffb1905d9d0543276b5f6ffb2d5cf29","source":{"kind":"arxiv","id":"2606.19399","version":1},"attestation_state":"computed","paper":{"title":"VERITAS: Verifier-Guided Proof Search for Zero-Shot Formal Theorem Proving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LO","cs.PL"],"primary_cat":"cs.LG","authors_text":"Kevin Leach, Manish Acharya, Yifan Zhang, Yueke Zhang, Yu Huang, Zhenyu Liao","submitted_at":"2026-06-17T09:08:39Z","abstract_excerpt":"LLM-based formal provers often collapse rich verifier signals (syntax errors, type mismatches, partial goal progress) into a binary pass/fail bit. We present VERITAS, a zero-shot framework that routes every verifier signal back into proof search through a two-phase protocol: Best-of-N sampling first, then a critic-guided MCTS pass that ingests Phase 1 failures as explicit negative examples. The protocol preserves every theorem solved by its own Phase 1 sweep, so Phase 2's additional solves are attributable to feedback-driven exploration. VERITAS reaches 40.6% on miniF2F (vs. an independently r"},"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":"2606.19399","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T09:08:39Z","cross_cats_sorted":["cs.AI","cs.LO","cs.PL"],"title_canon_sha256":"136f86254f40e250c4a9e3e2054af34d81c210552138b97db5d00eb1c618c1e9","abstract_canon_sha256":"944013dc8a4c487e763df6bc46d98edf13456b9cac0ec75478b470537d85f473"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:24.562682Z","signature_b64":"0ukx4WbM1CgiTb7Ot5lrAA4I3QO5oHjf4BuNfGWQPceF6rWlnGcDaaf59H0SmJuE665tLsFpl7wcLOPNEs5yCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7eb553ae9f18c49372e60908e113b584fffb1905d9d0543276b5f6ffb2d5cf29","last_reissued_at":"2026-06-19T16:12:24.562337Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:24.562337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VERITAS: Verifier-Guided Proof Search for Zero-Shot Formal Theorem Proving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LO","cs.PL"],"primary_cat":"cs.LG","authors_text":"Kevin Leach, Manish Acharya, Yifan Zhang, Yueke Zhang, Yu Huang, Zhenyu Liao","submitted_at":"2026-06-17T09:08:39Z","abstract_excerpt":"LLM-based formal provers often collapse rich verifier signals (syntax errors, type mismatches, partial goal progress) into a binary pass/fail bit. We present VERITAS, a zero-shot framework that routes every verifier signal back into proof search through a two-phase protocol: Best-of-N sampling first, then a critic-guided MCTS pass that ingests Phase 1 failures as explicit negative examples. The protocol preserves every theorem solved by its own Phase 1 sweep, so Phase 2's additional solves are attributable to feedback-driven exploration. VERITAS reaches 40.6% on miniF2F (vs. an independently r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19399","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/2606.19399/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":"2606.19399","created_at":"2026-06-19T16:12:24.562398+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.19399v1","created_at":"2026-06-19T16:12:24.562398+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19399","created_at":"2026-06-19T16:12:24.562398+00:00"},{"alias_kind":"pith_short_12","alias_value":"P22VHLU7DDCJ","created_at":"2026-06-19T16:12:24.562398+00:00"},{"alias_kind":"pith_short_16","alias_value":"P22VHLU7DDCJG4XG","created_at":"2026-06-19T16:12:24.562398+00:00"},{"alias_kind":"pith_short_8","alias_value":"P22VHLU7","created_at":"2026-06-19T16:12:24.562398+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/P22VHLU7DDCJG4XGBEEOCE5VQT","json":"https://pith.science/pith/P22VHLU7DDCJG4XGBEEOCE5VQT.json","graph_json":"https://pith.science/api/pith-number/P22VHLU7DDCJG4XGBEEOCE5VQT/graph.json","events_json":"https://pith.science/api/pith-number/P22VHLU7DDCJG4XGBEEOCE5VQT/events.json","paper":"https://pith.science/paper/P22VHLU7"},"agent_actions":{"view_html":"https://pith.science/pith/P22VHLU7DDCJG4XGBEEOCE5VQT","download_json":"https://pith.science/pith/P22VHLU7DDCJG4XGBEEOCE5VQT.json","view_paper":"https://pith.science/paper/P22VHLU7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.19399&json=true","fetch_graph":"https://pith.science/api/pith-number/P22VHLU7DDCJG4XGBEEOCE5VQT/graph.json","fetch_events":"https://pith.science/api/pith-number/P22VHLU7DDCJG4XGBEEOCE5VQT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P22VHLU7DDCJG4XGBEEOCE5VQT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P22VHLU7DDCJG4XGBEEOCE5VQT/action/storage_attestation","attest_author":"https://pith.science/pith/P22VHLU7DDCJG4XGBEEOCE5VQT/action/author_attestation","sign_citation":"https://pith.science/pith/P22VHLU7DDCJG4XGBEEOCE5VQT/action/citation_signature","submit_replication":"https://pith.science/pith/P22VHLU7DDCJG4XGBEEOCE5VQT/action/replication_record"}},"created_at":"2026-06-19T16:12:24.562398+00:00","updated_at":"2026-06-19T16:12:24.562398+00:00"}