{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6YFZ65ZAMXCLGE5X3QXE7MIZYY","short_pith_number":"pith:6YFZ65ZA","schema_version":"1.0","canonical_sha256":"f60b9f772065c4b313b7dc2e4fb119c63255043151fc4984d2f0aee56921d044","source":{"kind":"arxiv","id":"2509.26076","version":2},"attestation_state":"computed","paper":{"title":"IMProofBench: Benchmarking AI on Research-Level Mathematical Proof Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Adam Kurpisz, Aitor Iribar Lopez, Aleksandar Mijatovi\\'c, Alessio Bottini, Aluna Rizzoli, Ana Cannas da Silva, Ana-Maria Castravet, Baran Hashemi, Benjamin Doerr, Charles Vial, Chiara Meroni, Claudio Fontanari, Daniel Holmes, Daniel Platt, Daria Sakhanda, David Holmes, David Martins, David Mu\\~noz-Lahoz, Diaaeldin Taha, Dylan Possama\\\"i, Filippo Gaia, Gabriel Ribeiro, Georg Oberdieck, Gergely B\\'erczi, Henk van der Pol, Henry Liu, Ignacio Barros, Ingmar Metzler, Jasper Dekoninck, Jeremy Feusi, Jim Bryan, Jo\\~ao Camarneiro, Johannes Lengler, Johannes Schmitt, Josef Teichmann, Marc Roth, Martijn Kool, Martina J{\\o}rgensen, Martin M\\\"oller, Michel van Garrel, Niklas Canova, Patrick Schnider, Peter Gr\\\"unwald, Pieter Belmans, Pim Spelier, Raphael Appenzeller, Ra\\'ul S\\'anchez Gal\\'an, Richard P. Thomas, Robert Nowak, Ronald van Luijk, Samuel Mu\\~noz-Ech\\'aniz, Sergej Monavari, Stefan Kuhlmann, Steven Kelk, Tim Gehrunger, Timo de Wolff, Victor Jaeck, Yannik Schuler, Yuuji Tanaka, Zheming Sun","submitted_at":"2025-09-30T10:50:37Z","abstract_excerpt":"As the mathematical capabilities of large language models (LLMs) improve, it becomes increasingly important to evaluate their performance on research-level tasks at the frontier of mathematical knowledge. However, existing benchmarks are limited, as they focus solely on final-answer questions or high-school competition problems. To address this gap, we introduce IMProofBench, a private benchmark consisting of 77 peer-reviewed problems developed by expert mathematicians. Each problem requires a detailed proof and is paired with subproblems that have final answers, supporting both an evaluation "},"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":"2509.26076","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-30T10:50:37Z","cross_cats_sorted":[],"title_canon_sha256":"885e4921a7ac22b51c7b6b08347db14250690ce28767e8904b9b832bcecd6463","abstract_canon_sha256":"078632ad139486c8707c08ab5c8a0da2aa872f426f324b94164bac8927f66e48"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-10T01:19:39.853825Z","signature_b64":"r1APqUFgl+DkMn7tA6d1MHuQZJnjlw3NS3lGpRwDNjocH1VV2rzVKs0Q4Q8RUOXFlLVPvV3d5tuEIr1WRXsKBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f60b9f772065c4b313b7dc2e4fb119c63255043151fc4984d2f0aee56921d044","last_reissued_at":"2026-07-10T01:19:39.853324Z","signature_status":"signed_v1","first_computed_at":"2026-07-10T01:19:39.853324Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"IMProofBench: Benchmarking AI on Research-Level Mathematical Proof Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Adam Kurpisz, Aitor Iribar Lopez, Aleksandar Mijatovi\\'c, Alessio Bottini, Aluna Rizzoli, Ana Cannas da Silva, Ana-Maria Castravet, Baran Hashemi, Benjamin Doerr, Charles Vial, Chiara Meroni, Claudio Fontanari, Daniel Holmes, Daniel Platt, Daria Sakhanda, David Holmes, David Martins, David Mu\\~noz-Lahoz, Diaaeldin Taha, Dylan Possama\\\"i, Filippo Gaia, Gabriel Ribeiro, Georg Oberdieck, Gergely B\\'erczi, Henk van der Pol, Henry Liu, Ignacio Barros, Ingmar Metzler, Jasper Dekoninck, Jeremy Feusi, Jim Bryan, Jo\\~ao Camarneiro, Johannes Lengler, Johannes Schmitt, Josef Teichmann, Marc Roth, Martijn Kool, Martina J{\\o}rgensen, Martin M\\\"oller, Michel van Garrel, Niklas Canova, Patrick Schnider, Peter Gr\\\"unwald, Pieter Belmans, Pim Spelier, Raphael Appenzeller, Ra\\'ul S\\'anchez Gal\\'an, Richard P. Thomas, Robert Nowak, Ronald van Luijk, Samuel Mu\\~noz-Ech\\'aniz, Sergej Monavari, Stefan Kuhlmann, Steven Kelk, Tim Gehrunger, Timo de Wolff, Victor Jaeck, Yannik Schuler, Yuuji Tanaka, Zheming Sun","submitted_at":"2025-09-30T10:50:37Z","abstract_excerpt":"As the mathematical capabilities of large language models (LLMs) improve, it becomes increasingly important to evaluate their performance on research-level tasks at the frontier of mathematical knowledge. However, existing benchmarks are limited, as they focus solely on final-answer questions or high-school competition problems. To address this gap, we introduce IMProofBench, a private benchmark consisting of 77 peer-reviewed problems developed by expert mathematicians. Each problem requires a detailed proof and is paired with subproblems that have final answers, supporting both an evaluation "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.26076","kind":"arxiv","version":2},"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/2509.26076/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":"2509.26076","created_at":"2026-07-10T01:19:39.853380+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.26076v2","created_at":"2026-07-10T01:19:39.853380+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.26076","created_at":"2026-07-10T01:19:39.853380+00:00"},{"alias_kind":"pith_short_12","alias_value":"6YFZ65ZAMXCL","created_at":"2026-07-10T01:19:39.853380+00:00"},{"alias_kind":"pith_short_16","alias_value":"6YFZ65ZAMXCLGE5X","created_at":"2026-07-10T01:19:39.853380+00:00"},{"alias_kind":"pith_short_8","alias_value":"6YFZ65ZA","created_at":"2026-07-10T01:19:39.853380+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":6,"sample":[{"citing_arxiv_id":"2607.06820","citing_title":"Evaluating SageMath-Augmented LLM Agents for Computational and Experimental Mathematics","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2606.24902","citing_title":"Failure Modes of Large Language Models on Research-Level Mathematics: A Taxonomy and an Empirical Characterisation","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10379","citing_title":"Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09063","citing_title":"Soohak: A Mathematician-Curated Benchmark for Evaluating Research-level Math Capabilities of LLMs","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09063","citing_title":"Soohak: A Mathematician-Curated Benchmark for Evaluating Research-level Math Capabilities of LLMs","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10379","citing_title":"Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY","json":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY.json","graph_json":"https://pith.science/api/pith-number/6YFZ65ZAMXCLGE5X3QXE7MIZYY/graph.json","events_json":"https://pith.science/api/pith-number/6YFZ65ZAMXCLGE5X3QXE7MIZYY/events.json","paper":"https://pith.science/paper/6YFZ65ZA"},"agent_actions":{"view_html":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY","download_json":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY.json","view_paper":"https://pith.science/paper/6YFZ65ZA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.26076&json=true","fetch_graph":"https://pith.science/api/pith-number/6YFZ65ZAMXCLGE5X3QXE7MIZYY/graph.json","fetch_events":"https://pith.science/api/pith-number/6YFZ65ZAMXCLGE5X3QXE7MIZYY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY/action/storage_attestation","attest_author":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY/action/author_attestation","sign_citation":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY/action/citation_signature","submit_replication":"https://pith.science/pith/6YFZ65ZAMXCLGE5X3QXE7MIZYY/action/replication_record"}},"created_at":"2026-07-10T01:19:39.853380+00:00","updated_at":"2026-07-10T01:19:39.853380+00:00"}