{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AXM3Z5VQ4GSXUGKF4PYAE72ULC","short_pith_number":"pith:AXM3Z5VQ","schema_version":"1.0","canonical_sha256":"05d9bcf6b0e1a57a1945e3f0027f5458a157336e6bd2f7204b49ed815a25ce96","source":{"kind":"arxiv","id":"2508.03613","version":1},"attestation_state":"computed","paper":{"title":"Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bohan Lyu, Chi Jin, Danqi Chen, David Acuna, Haoyu Zhao, Hongzhou Lin, Jiawei Ge, Jiayun Wu, Jingruo Sun, Jiri Gesi, Jui-Hui Chung, Kaiyu Yang, Lai Jiang, Sanjeev Arora, Shange Tang, Ximing Lu, Yejin Choi, Yihan Geng, Yong Lin, Ziran Yang","submitted_at":"2025-08-05T16:28:22Z","abstract_excerpt":"We introduce Goedel-Prover-V2, a series of open-source language models that set a new state-of-the-art in automated theorem proving. Built on the standard expert iteration and reinforcement learning pipeline, our approach incorporates three key innovations: (1) Scaffolded data synthesis: We generate synthetic tasks of increasing difficulty to train the model to master increasingly complex theorems; (2) Verifier-guided self-correction: We enable the model to iteratively revise its proofs by leveraging feedback from the Lean compiler; (3) Model averaging: We merge model checkpoints to mitigate t"},"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":"2508.03613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-05T16:28:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b5e7c87057a3afe7c2ed58cc68650ab28aef379d698222b710ec7bb4b9b69735","abstract_canon_sha256":"09101f5f879db5260b1515a46d1edeb12dc0aa6500a7132c0a6f146648791ffb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T06:47:44.712882Z","signature_b64":"TuHRup6keib5kXs5Lvs3o1WMoIpWkK9hO/CiZKbpga+jW4Cr8w0Q+pSFKnKoQXNa+avT3HHF38x6RNO5w1w7AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"05d9bcf6b0e1a57a1945e3f0027f5458a157336e6bd2f7204b49ed815a25ce96","last_reissued_at":"2026-05-21T06:47:44.711218Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T06:47:44.711218Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bohan Lyu, Chi Jin, Danqi Chen, David Acuna, Haoyu Zhao, Hongzhou Lin, Jiawei Ge, Jiayun Wu, Jingruo Sun, Jiri Gesi, Jui-Hui Chung, Kaiyu Yang, Lai Jiang, Sanjeev Arora, Shange Tang, Ximing Lu, Yejin Choi, Yihan Geng, Yong Lin, Ziran Yang","submitted_at":"2025-08-05T16:28:22Z","abstract_excerpt":"We introduce Goedel-Prover-V2, a series of open-source language models that set a new state-of-the-art in automated theorem proving. Built on the standard expert iteration and reinforcement learning pipeline, our approach incorporates three key innovations: (1) Scaffolded data synthesis: We generate synthetic tasks of increasing difficulty to train the model to master increasingly complex theorems; (2) Verifier-guided self-correction: We enable the model to iteratively revise its proofs by leveraging feedback from the Lean compiler; (3) Model averaging: We merge model checkpoints to mitigate t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.03613","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/2508.03613/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":"2508.03613","created_at":"2026-05-21T06:47:44.711272+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.03613v1","created_at":"2026-05-21T06:47:44.711272+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.03613","created_at":"2026-05-21T06:47:44.711272+00:00"},{"alias_kind":"pith_short_12","alias_value":"AXM3Z5VQ4GSX","created_at":"2026-05-21T06:47:44.711272+00:00"},{"alias_kind":"pith_short_16","alias_value":"AXM3Z5VQ4GSXUGKF","created_at":"2026-05-21T06:47:44.711272+00:00"},{"alias_kind":"pith_short_8","alias_value":"AXM3Z5VQ","created_at":"2026-05-21T06:47:44.711272+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":24,"internal_anchor_count":24,"sample":[{"citing_arxiv_id":"2605.22763","citing_title":"Advancing Mathematics Research with AI-Driven Formal Proof Search","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20531","citing_title":"Pseudo-Formalization for Automatic Proof Verification","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17283","citing_title":"OProver: A Unified Framework for Agentic Formal Theorem Proving","ref_index":172,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17255","citing_title":"CAM-Bench: A Benchmark for Computational and Applied Mathematics in Lean","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18747","citing_title":"Code as Agent Harness","ref_index":89,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17778","citing_title":"Self-Distillation is Optimal Among Spectral Shrinkage Estimators in Spiked Covariance Models","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2509.14274","citing_title":"Discovering New Theorems via LLMs with In-Context Proof Learning in Lean","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2602.24273","citing_title":"A Minimal Agent for Automated Theorem Proving","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14061","citing_title":"MathAtlas: A Benchmark for Autoformalization in the Wild","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06651","citing_title":"AI co-mathematician: Accelerating mathematicians with agentic AI","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2604.03071","citing_title":"Automatic Textbook Formalization","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11905","citing_title":"Rethinking Supervision Granularity: Segment-Level Learning for LLM-Based Theorem Proving","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08678","citing_title":"MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10379","citing_title":"Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2604.23712","citing_title":"OptProver: Bridging Olympiad and Optimization through Continual Training in Formal Theorem Proving","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06651","citing_title":"AI co-mathematician: Accelerating mathematicians with agentic AI","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22519","citing_title":"Ablation and the Meno: Tools for Empirical Metamathematics","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06110","citing_title":"On Time, Within Budget: Constraint-Driven Online Resource Allocation for Agentic Workflows","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00677","citing_title":"Evaluating the Architectural Reasoning Capabilities of LLM Provers via the Obfuscated Natural Number Game","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20209","citing_title":"Scaling Self-Play with Self-Guidance","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19558","citing_title":"On Reasoning-Centric LLM-based Automated Theorem Proving","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06110","citing_title":"On Time, Within Budget: Constraint-Driven Online Resource Allocation for Agentic Workflows","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.02909","citing_title":"Delay, Plateau, or Collapse: Evaluating the Impact of Systematic Verification Error on RLVR","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2604.18050","citing_title":"The Topological Dual of a Dataset: A Logic-to-Topology Encoding for AlphaGeometry-Style Data","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC","json":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC.json","graph_json":"https://pith.science/api/pith-number/AXM3Z5VQ4GSXUGKF4PYAE72ULC/graph.json","events_json":"https://pith.science/api/pith-number/AXM3Z5VQ4GSXUGKF4PYAE72ULC/events.json","paper":"https://pith.science/paper/AXM3Z5VQ"},"agent_actions":{"view_html":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC","download_json":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC.json","view_paper":"https://pith.science/paper/AXM3Z5VQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.03613&json=true","fetch_graph":"https://pith.science/api/pith-number/AXM3Z5VQ4GSXUGKF4PYAE72ULC/graph.json","fetch_events":"https://pith.science/api/pith-number/AXM3Z5VQ4GSXUGKF4PYAE72ULC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC/action/storage_attestation","attest_author":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC/action/author_attestation","sign_citation":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC/action/citation_signature","submit_replication":"https://pith.science/pith/AXM3Z5VQ4GSXUGKF4PYAE72ULC/action/replication_record"}},"created_at":"2026-05-21T06:47:44.711272+00:00","updated_at":"2026-05-21T06:47:44.711272+00:00"}