{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WK7MJNIPW5OLCMQ5UQJEJKG3SB","short_pith_number":"pith:WK7MJNIP","schema_version":"1.0","canonical_sha256":"b2bec4b50fb75cb1321da41244a8db904d0ce5e42588aed9b19845e181eeafe3","source":{"kind":"arxiv","id":"2503.21934","version":5},"attestation_state":"computed","paper":{"title":"Proof or Bluff? Evaluating LLMs on 2025 USA Math Olympiad","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ivo Petrov, Jasper Dekoninck, Kristian Minchev, Lyuben Baltadzhiev, Maria Drencheva, Martin Vechev, Mislav Balunovi\\'c, Nikola Jovanovi\\'c","submitted_at":"2025-03-27T19:21:05Z","abstract_excerpt":"Recent math benchmarks for large language models (LLMs) such as MathArena indicate that state-of-the-art reasoning models achieve impressive performance on mathematical competitions like AIME, with the leading model, Gemini-2.5-Pro, achieving scores comparable to top human competitors. However, these benchmarks evaluate models solely based on final numerical answers, neglecting rigorous reasoning and proof generation which are essential for real-world mathematical tasks. To address this, we introduce a comprehensive evaluation of full-solution reasoning for challenging mathematical problems. U"},"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":"2503.21934","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-27T19:21:05Z","cross_cats_sorted":[],"title_canon_sha256":"604fc6d644092207ef784b2a059fdd834f71710b9fe57a0c64d9700f6f609a61","abstract_canon_sha256":"aa3fa19ababd8784465f7f7ca2ada7a0c1aa52d44bff08d07df648d7a0512277"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:15.336475Z","signature_b64":"QzDtFF+EkZLT5ru6dW+KJR+0ftXFrnnCgSacq58E19vQis2WU/o/SOcaBTs7e48lUmSFMpMkZbkacGip9f6WDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2bec4b50fb75cb1321da41244a8db904d0ce5e42588aed9b19845e181eeafe3","last_reissued_at":"2026-07-05T12:05:15.335885Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:15.335885Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Proof or Bluff? Evaluating LLMs on 2025 USA Math Olympiad","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ivo Petrov, Jasper Dekoninck, Kristian Minchev, Lyuben Baltadzhiev, Maria Drencheva, Martin Vechev, Mislav Balunovi\\'c, Nikola Jovanovi\\'c","submitted_at":"2025-03-27T19:21:05Z","abstract_excerpt":"Recent math benchmarks for large language models (LLMs) such as MathArena indicate that state-of-the-art reasoning models achieve impressive performance on mathematical competitions like AIME, with the leading model, Gemini-2.5-Pro, achieving scores comparable to top human competitors. However, these benchmarks evaluate models solely based on final numerical answers, neglecting rigorous reasoning and proof generation which are essential for real-world mathematical tasks. To address this, we introduce a comprehensive evaluation of full-solution reasoning for challenging mathematical problems. U"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21934","kind":"arxiv","version":5},"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/2503.21934/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":"2503.21934","created_at":"2026-07-05T12:05:15.335949+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.21934v5","created_at":"2026-07-05T12:05:15.335949+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21934","created_at":"2026-07-05T12:05:15.335949+00:00"},{"alias_kind":"pith_short_12","alias_value":"WK7MJNIPW5OL","created_at":"2026-07-05T12:05:15.335949+00:00"},{"alias_kind":"pith_short_16","alias_value":"WK7MJNIPW5OLCMQ5","created_at":"2026-07-05T12:05:15.335949+00:00"},{"alias_kind":"pith_short_8","alias_value":"WK7MJNIP","created_at":"2026-07-05T12:05:15.335949+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24020","citing_title":"You Don't Need to Run Every Eval","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10379","citing_title":"Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2504.13217","citing_title":"Sustainability via LLM Right-sizing","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2509.21820","citing_title":"Can LLMs Generate and Solve Linguistic Olympiad Puzzles?","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19338","citing_title":"STAR-P\\'olyaMath: Multi-Agent Reasoning under Persistent Meta-Strategic Supervision","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2511.02521","citing_title":"Large Lemma Miners: Can LLMs do Induction Proofs for Hardware?","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2511.10899","citing_title":"From Proof to Program: Characterizing Tool-Induced Reasoning Hallucinations in Large Language Models","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2504.21318","citing_title":"Phi-4-reasoning Technical Report","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2505.23281","citing_title":"MathArena: Evaluating LLMs on Uncontaminated Math Competitions","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03216","citing_title":"BAS: A Decision-Theoretic Approach to Evaluating Large Language Model Confidence","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10379","citing_title":"Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07593","citing_title":"Too long; didn't solve","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB","json":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB.json","graph_json":"https://pith.science/api/pith-number/WK7MJNIPW5OLCMQ5UQJEJKG3SB/graph.json","events_json":"https://pith.science/api/pith-number/WK7MJNIPW5OLCMQ5UQJEJKG3SB/events.json","paper":"https://pith.science/paper/WK7MJNIP"},"agent_actions":{"view_html":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB","download_json":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB.json","view_paper":"https://pith.science/paper/WK7MJNIP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.21934&json=true","fetch_graph":"https://pith.science/api/pith-number/WK7MJNIPW5OLCMQ5UQJEJKG3SB/graph.json","fetch_events":"https://pith.science/api/pith-number/WK7MJNIPW5OLCMQ5UQJEJKG3SB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB/action/storage_attestation","attest_author":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB/action/author_attestation","sign_citation":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB/action/citation_signature","submit_replication":"https://pith.science/pith/WK7MJNIPW5OLCMQ5UQJEJKG3SB/action/replication_record"}},"created_at":"2026-07-05T12:05:15.335949+00:00","updated_at":"2026-07-05T12:05:15.335949+00:00"}