{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KUFHIHH4IF4BZMYVMXRN37C5JO","short_pith_number":"pith:KUFHIHH4","schema_version":"1.0","canonical_sha256":"550a741cfc41781cb31565e2ddfc5d4bb3fc1167ab9b3ddcfc8ea92ff58eaf57","source":{"kind":"arxiv","id":"2605.00674","version":2},"attestation_state":"computed","paper":{"title":"Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Expanding MathArena to proofs and research questions shows frontier LLMs solve 98% of 2026 USAMO problems.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenhao Sun, Ivo Petrov, Jasper Dekoninck, K\\'ari R\\\"ognvaldsson, Martin Vechev, Nikola Jovanovi\\'c, Tim Gehrunger","submitted_at":"2026-05-01T13:56:34Z","abstract_excerpt":"Large language models (LLMs) are becoming increasingly capable mathematical collaborators, but static benchmarks are no longer sufficient for evaluating progress: they are often narrow in scope, quickly saturated, and rarely updated. This makes it hard to compare models reliably and track progress over time. Instead, we need evaluation platforms: continuously maintained systems that run, aggregate, and analyze evaluations across many benchmarks to give a comprehensive picture of model performance within a broad domain. In this work, we build on the original MathArena benchmark by substantially"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.00674","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-01T13:56:34Z","cross_cats_sorted":[],"title_canon_sha256":"81650e89f738211c5de46e0d9f3372b37e7c2953c81b63a0d9e9f19156d4c92d","abstract_canon_sha256":"f2de4624f6470c58c9995a5f89759c13962d99f1460444f83ddc681705973ef1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:42.393359Z","signature_b64":"WKSSf03R35vZUzhEpQ0QE5ce/2SxZrwqepPyq3zHhritqIxsqcAM7+DeJrQH4PGn/UpWbRApcGVs6qrXCLs5DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"550a741cfc41781cb31565e2ddfc5d4bb3fc1167ab9b3ddcfc8ea92ff58eaf57","last_reissued_at":"2026-05-20T00:01:42.392550Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:42.392550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Expanding MathArena to proofs and research questions shows frontier LLMs solve 98% of 2026 USAMO problems.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenhao Sun, Ivo Petrov, Jasper Dekoninck, K\\'ari R\\\"ognvaldsson, Martin Vechev, Nikola Jovanovi\\'c, Tim Gehrunger","submitted_at":"2026-05-01T13:56:34Z","abstract_excerpt":"Large language models (LLMs) are becoming increasingly capable mathematical collaborators, but static benchmarks are no longer sufficient for evaluating progress: they are often narrow in scope, quickly saturated, and rarely updated. This makes it hard to compare models reliably and track progress over time. Instead, we need evaluation platforms: continuously maintained systems that run, aggregate, and analyze evaluations across many benchmarks to give a comprehensive picture of model performance within a broad domain. In this work, we build on the original MathArena benchmark by substantially"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Notably, the strongest model, GPT-5.5, now reaches 98% on the 2026 USA Math Olympiad and 74% on research-level questions, showing that frontier models can now comfortably solve extremely challenging mathematical problems.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The new benchmarks and evaluation protocol remain free of training-data contamination and fairly measure genuine reasoning rather than memorization or benchmark-specific shortcuts.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"MathArena is broadened into a maintained platform with new benchmarks for proofs, research questions, and formal verification, where GPT-5.5 scores 98% on 2026 USAMO and 74% on research-level tasks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Expanding MathArena to proofs and research questions shows frontier LLMs solve 98% of 2026 USAMO problems.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"14dd8eeb1bd2b60189b733bf4f60f2d3ce5085527627ef0f4d88d12607437aba"},"source":{"id":"2605.00674","kind":"arxiv","version":2},"verdict":{"id":"b18e07f0-5273-4711-9f83-a4ed862bd3e1","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T18:14:31.031764Z","strongest_claim":"Notably, the strongest model, GPT-5.5, now reaches 98% on the 2026 USA Math Olympiad and 74% on research-level questions, showing that frontier models can now comfortably solve extremely challenging mathematical problems.","one_line_summary":"MathArena is broadened into a maintained platform with new benchmarks for proofs, research questions, and formal verification, where GPT-5.5 scores 98% on 2026 USAMO and 74% on research-level tasks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The new benchmarks and evaluation protocol remain free of training-data contamination and fairly measure genuine reasoning rather than memorization or benchmark-specific shortcuts.","pith_extraction_headline":"Expanding MathArena to proofs and research questions shows frontier LLMs solve 98% of 2026 USAMO problems."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.00674/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T17:56:14.716195Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"b7ed88aa31ffa30aa3052fb1d4cc4d988fa53280d5bfc7f48f2d9f1c5eccc211"},"references":{"count":110,"sample":[{"doi":"","year":2024,"title":"Wei-Lin Chiang, Lianmin Zheng, Ying Sheng, Anastasios Nikolas Angelopoulos, Tianle Li, Dacheng Li, Banghua Zhu, Hao Zhang, Michael I. Jordan, Joseph E. Gonzalez, and Ion Stoica. Chatbot arena: An open","work_id":"b113a7e2-dc84-4564-817e-141ee365f5a6","ref_index":6,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Ai model & api providers analysis","work_id":"e6067ebc-31e4-4f46-aa93-c2b894abb00d","ref_index":7,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"Accessed: 2026-04-13","work_id":"69f838f3-378d-4111-9988-826f26d14ec3","ref_index":8,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"Vals AI. Benchmarks.https://www.vals.ai/benchmarks, 2026. Accessed: 2026-04-13","work_id":"fc94e773-4ed5-4745-861e-5c6462ba401f","ref_index":9,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"MathArena: Evaluating LLMs on Uncontaminated Math Competitions","work_id":"61e8d872-ccc7-46b8-8ec1-94008704c941","ref_index":10,"cited_arxiv_id":"2505.23281","is_internal_anchor":true}],"resolved_work":110,"snapshot_sha256":"3e572e08799c52073806386ff2a33bd44ca3b6414a388a1558896f98b81125b2","internal_anchors":5},"formal_canon":{"evidence_count":2,"snapshot_sha256":"d8ae847744fa30b19249002cc1fb50bccad3fedb85cb8ae12e2c51852efaf8e3"},"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":"2605.00674","created_at":"2026-05-20T00:01:42.392701+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.00674v2","created_at":"2026-05-20T00:01:42.392701+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.00674","created_at":"2026-05-20T00:01:42.392701+00:00"},{"alias_kind":"pith_short_12","alias_value":"KUFHIHH4IF4B","created_at":"2026-05-20T00:01:42.392701+00:00"},{"alias_kind":"pith_short_16","alias_value":"KUFHIHH4IF4BZMYV","created_at":"2026-05-20T00:01:42.392701+00:00"},{"alias_kind":"pith_short_8","alias_value":"KUFHIHH4","created_at":"2026-05-20T00:01:42.392701+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":9,"sample":[{"citing_arxiv_id":"2605.21606","citing_title":"When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20315","citing_title":"Mix-Quant: Quantized Prefilling, Precise Decoding for Agentic LLMs","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21468","citing_title":"You Only Need Minimal RLVR Training: Extrapolating LLMs via Rank-1 Trajectories","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21008","citing_title":"A Survey of Audio Reasoning in Multimodal Foundation Models","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11905","citing_title":"Rethinking Supervision Granularity: Segment-Level Learning for LLM-Based Theorem Proving","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12193","citing_title":"BFLA: Block-Filtered Long-Context Attention Mechanism","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11609","citing_title":"Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05873","citing_title":"CITE: Anytime-Valid Statistical Inference in LLM Self-Consistency","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06326","citing_title":"Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated Reasoning","ref_index":7,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO","json":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO.json","graph_json":"https://pith.science/api/pith-number/KUFHIHH4IF4BZMYVMXRN37C5JO/graph.json","events_json":"https://pith.science/api/pith-number/KUFHIHH4IF4BZMYVMXRN37C5JO/events.json","paper":"https://pith.science/paper/KUFHIHH4"},"agent_actions":{"view_html":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO","download_json":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO.json","view_paper":"https://pith.science/paper/KUFHIHH4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.00674&json=true","fetch_graph":"https://pith.science/api/pith-number/KUFHIHH4IF4BZMYVMXRN37C5JO/graph.json","fetch_events":"https://pith.science/api/pith-number/KUFHIHH4IF4BZMYVMXRN37C5JO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO/action/storage_attestation","attest_author":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO/action/author_attestation","sign_citation":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO/action/citation_signature","submit_replication":"https://pith.science/pith/KUFHIHH4IF4BZMYVMXRN37C5JO/action/replication_record"}},"created_at":"2026-05-20T00:01:42.392701+00:00","updated_at":"2026-05-20T00:01:42.392701+00:00"}