{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7DMISRS7T2TMSVT23ZQWMYQBL7","short_pith_number":"pith:7DMISRS7","schema_version":"1.0","canonical_sha256":"f8d889465f9ea6c9567ade616662015fcf8899325d78cadaaa64e4f713039c33","source":{"kind":"arxiv","id":"2504.11456","version":2},"attestation_state":"computed","paper":{"title":"DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"DeepMath-103K supplies 103K hard, clean math problems that let reinforcement learning reach state-of-the-art reasoning performance.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dian Yu, Dong Yu, Haitao Mi, Jiahao Xu, Linfeng Song, Qiuzhi Liu, Rui Wang, Tian Liang, Wenxuan Wang, Xingyu Chen, Yue Wang, Zhaopeng Tu, Zhenwen Liang, Zhiwei He, Zhuosheng Zhang","submitted_at":"2025-04-15T17:59:51Z","abstract_excerpt":"Reinforcement learning (RL) with large language models shows promise in complex reasoning. However, its progress is hindered by the lack of large-scale training data that is sufficiently challenging, contamination-free and verifiable. To this end, we introduce DeepMath-103K, a large-scale mathematical dataset designed with high difficulty (primarily levels 5-9), rigorous decontamination against numerous benchmarks, and verifiable answers for rule-based RL reward. It further includes three distinct R1 solutions adaptable for diverse training paradigms such as supervised fine-tuning (SFT). Spann"},"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":"2504.11456","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T17:59:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"53e33e5e58a1bb68b4dc91d0f24d06d1a094d9a28713b715d476eeb3c32a2cf6","abstract_canon_sha256":"07c276a3630efe36049efdcd5a7b0393561f3daaeb3ad81157d140ec6a7b5b34"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:48.188571Z","signature_b64":"J/wbu5N2RpLYetNbRIGl6L+ZLKY5+puxlR0eH43P/C+w5KAmoibdsaT1vocthz4ttHI7K+E8EyMZuMGbMubDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8d889465f9ea6c9567ade616662015fcf8899325d78cadaaa64e4f713039c33","last_reissued_at":"2026-05-17T23:38:48.188069Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:48.188069Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"DeepMath-103K supplies 103K hard, clean math problems that let reinforcement learning reach state-of-the-art reasoning performance.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dian Yu, Dong Yu, Haitao Mi, Jiahao Xu, Linfeng Song, Qiuzhi Liu, Rui Wang, Tian Liang, Wenxuan Wang, Xingyu Chen, Yue Wang, Zhaopeng Tu, Zhenwen Liang, Zhiwei He, Zhuosheng Zhang","submitted_at":"2025-04-15T17:59:51Z","abstract_excerpt":"Reinforcement learning (RL) with large language models shows promise in complex reasoning. However, its progress is hindered by the lack of large-scale training data that is sufficiently challenging, contamination-free and verifiable. To this end, we introduce DeepMath-103K, a large-scale mathematical dataset designed with high difficulty (primarily levels 5-9), rigorous decontamination against numerous benchmarks, and verifiable answers for rule-based RL reward. It further includes three distinct R1 solutions adaptable for diverse training paradigms such as supervised fine-tuning (SFT). Spann"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"models trained on DeepMath-103K achieve state-of-the-art results on challenging mathematical benchmarks and demonstrate generalization beyond math such as biology, physics and chemistry","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The decontamination process fully removes overlap with numerous benchmarks and the selected problems remain sufficiently challenging and verifiable to produce genuine gains in reasoning capability.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DeepMath-103K is a new 103K-problem mathematical dataset with high difficulty, rigorous decontamination, and verifiable answers to support RL training of language-model reasoning.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"DeepMath-103K supplies 103K hard, clean math problems that let reinforcement learning reach state-of-the-art reasoning performance.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"dd32e69dd7a3534871410530c6b7658000018536302e28df13b4762017b619b8"},"source":{"id":"2504.11456","kind":"arxiv","version":2},"verdict":{"id":"ee96a8e9-1767-4fb8-93f7-e4cf7534a95e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T10:27:45.145056Z","strongest_claim":"models trained on DeepMath-103K achieve state-of-the-art results on challenging mathematical benchmarks and demonstrate generalization beyond math such as biology, physics and chemistry","one_line_summary":"DeepMath-103K is a new 103K-problem mathematical dataset with high difficulty, rigorous decontamination, and verifiable answers to support RL training of language-model reasoning.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The decontamination process fully removes overlap with numerous benchmarks and the selected problems remain sufficiently challenging and verifiable to produce genuine gains in reasoning capability.","pith_extraction_headline":"DeepMath-103K supplies 103K hard, clean math problems that let reinforcement learning reach state-of-the-art reasoning performance."},"references":{"count":22,"sample":[{"doi":"","year":null,"title":"Marthe Ballon, Brecht Verbeken, Vincent Ginis, and Andres Algaba","work_id":"00edb0c5-3c62-421d-a532-03e2594f1dba","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model","work_id":"8472f581-14d4-40f8-8189-62ed8b470c4e","ref_index":2,"cited_arxiv_id":"2502.02737","is_internal_anchor":true},{"doi":"10.18653/v1/2023.emnlp-main.468","year":2023,"title":"doi: 10.18653/v1/2023.emnlp-main.468","work_id":"8d54fadf-29e9-49ff-aab2-e9bc4679ef97","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs","work_id":"d79f8b7d-560d-4fbd-bd70-4d084f6d9ad6","ref_index":4,"cited_arxiv_id":"2412.21187","is_internal_anchor":true},{"doi":"","year":null,"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","ref_index":5,"cited_arxiv_id":"2110.14168","is_internal_anchor":true}],"resolved_work":22,"snapshot_sha256":"1a1ffd0dff706fbc80115a2d8f45ae0744dca908e7641f8ffa625ad4cf6fe554","internal_anchors":11},"formal_canon":{"evidence_count":1,"snapshot_sha256":"cfecafb5d9f5c9af3dc6964d79e60f52a9401c53c81e8262383061d23716aa6f"},"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":"2504.11456","created_at":"2026-05-17T23:38:48.188152+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.11456v2","created_at":"2026-05-17T23:38:48.188152+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.11456","created_at":"2026-05-17T23:38:48.188152+00:00"},{"alias_kind":"pith_short_12","alias_value":"7DMISRS7T2TM","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"7DMISRS7T2TMSVT2","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"7DMISRS7","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":26,"internal_anchor_count":26,"sample":[{"citing_arxiv_id":"2605.08401","citing_title":"AIPO: Learning to Reason from Active Interaction","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2503.17352","citing_title":"OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2509.08827","citing_title":"A Survey of Reinforcement Learning for Large Reasoning Models","ref_index":193,"is_internal_anchor":true},{"citing_arxiv_id":"2511.20814","citing_title":"SPHINX: A Synthetic Environment for Visual Perception and Reasoning","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2601.04809","citing_title":"SCALER:Synthetic Scalable Adaptive Learning Environment for Reasoning","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2601.21619","citing_title":"On the Overscaling Curse of Parallel Thinking: System Efficacy Contradicts Sample Efficiency","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2601.21464","citing_title":"Conversation for Non-verifiable Learning: Self-Evolving LLMs through Meta-Evaluation","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12652","citing_title":"Multi-Rollout On-Policy Distillation via Peer Successes and Failures","ref_index":73,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06387","citing_title":"Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12492","citing_title":"Pion: A Spectrum-Preserving Optimizer via Orthogonal Equivalence Transformation","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11625","citing_title":"Nice Fold or Hero Call: Learning Budget-Efficient Thinking for Adaptive Reasoning","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27039","citing_title":"Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08472","citing_title":"Mid-Training with Self-Generated Data Improves Reinforcement Learning in Language Models","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08401","citing_title":"AIPO: Learning to Reason from Active Interaction","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08741","citing_title":"Training with Harnesses: On-Policy Harness Self-Distillation for Complex Reasoning","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09188","citing_title":"DARE: Difficulty-Adaptive Reinforcement Learning with Co-Evolved Difficulty Estimation","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08905","citing_title":"Forge: Quality-Aware Reinforcement Learning for NP-Hard Optimization in LLMs","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06523","citing_title":"On the Implicit Reward Overfitting and the Low-rank Dynamics in RLVR","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06387","citing_title":"Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10827","citing_title":"Your Model Diversity, Not Method, Determines Reasoning Strategy","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2604.12160","citing_title":"PubSwap: Public-Data Off-Policy Coordination for Federated RLVR","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06387","citing_title":"Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2604.18381","citing_title":"Learning from Less: Measuring the Effectiveness of RLVR in Low Data and Compute Regimes","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05226","citing_title":"Internalizing Outcome Supervision into Process Supervision: A New Paradigm for Reinforcement Learning for Reasoning","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17328","citing_title":"Rethinking the Comparison Unit in Sequence-Level Reinforcement Learning: An Equal-Length Paired Training Framework from Loss Correction to Sample Construction","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":1,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7","json":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7.json","graph_json":"https://pith.science/api/pith-number/7DMISRS7T2TMSVT23ZQWMYQBL7/graph.json","events_json":"https://pith.science/api/pith-number/7DMISRS7T2TMSVT23ZQWMYQBL7/events.json","paper":"https://pith.science/paper/7DMISRS7"},"agent_actions":{"view_html":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7","download_json":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7.json","view_paper":"https://pith.science/paper/7DMISRS7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.11456&json=true","fetch_graph":"https://pith.science/api/pith-number/7DMISRS7T2TMSVT23ZQWMYQBL7/graph.json","fetch_events":"https://pith.science/api/pith-number/7DMISRS7T2TMSVT23ZQWMYQBL7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7/action/storage_attestation","attest_author":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7/action/author_attestation","sign_citation":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7/action/citation_signature","submit_replication":"https://pith.science/pith/7DMISRS7T2TMSVT23ZQWMYQBL7/action/replication_record"}},"created_at":"2026-05-17T23:38:48.188152+00:00","updated_at":"2026-05-17T23:38:48.188152+00:00"}