{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7DMISRS7T2TMSVT23ZQWMYQBL7","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"07c276a3630efe36049efdcd5a7b0393561f3daaeb3ad81157d140ec6a7b5b34","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T17:59:51Z","title_canon_sha256":"53e33e5e58a1bb68b4dc91d0f24d06d1a094d9a28713b715d476eeb3c32a2cf6"},"schema_version":"1.0","source":{"id":"2504.11456","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.11456","created_at":"2026-05-17T23:38:48Z"},{"alias_kind":"arxiv_version","alias_value":"2504.11456v2","created_at":"2026-05-17T23:38:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.11456","created_at":"2026-05-17T23:38:48Z"},{"alias_kind":"pith_short_12","alias_value":"7DMISRS7T2TM","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"7DMISRS7T2TMSVT2","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"7DMISRS7","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:03dfb9e7aa746cf4d0b5cba7e1bf77889752bb79de14b0c9c35b9ff99d79644f","target":"graph","created_at":"2026-05-17T23:38:48Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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"},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"DeepMath-103K supplies 103K hard, clean math problems that let reinforcement learning reach state-of-the-art reasoning performance."}],"snapshot_sha256":"dd32e69dd7a3534871410530c6b7658000018536302e28df13b4762017b619b8"},"formal_canon":{"evidence_count":1,"snapshot_sha256":"cfecafb5d9f5c9af3dc6964d79e60f52a9401c53c81e8262383061d23716aa6f"},"paper":{"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","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","cross_cats":["cs.AI"],"headline":"DeepMath-103K supplies 103K hard, clean math problems that let reinforcement learning reach state-of-the-art reasoning performance.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T17:59:51Z","title":"DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning"},"references":{"count":22,"internal_anchors":11,"resolved_work":22,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Marthe Ballon, Brecht Verbeken, Vincent Ginis, and Andres Algaba","work_id":"00edb0c5-3c62-421d-a532-03e2594f1dba","year":null},{"cited_arxiv_id":"2502.02737","doi":"","is_internal_anchor":true,"ref_index":2,"title":"SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model","work_id":"8472f581-14d4-40f8-8189-62ed8b470c4e","year":2023},{"cited_arxiv_id":"","doi":"10.18653/v1/2023.emnlp-main.468","is_internal_anchor":false,"ref_index":3,"title":"doi: 10.18653/v1/2023.emnlp-main.468","work_id":"8d54fadf-29e9-49ff-aab2-e9bc4679ef97","year":2023},{"cited_arxiv_id":"2412.21187","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs","work_id":"d79f8b7d-560d-4fbd-bd70-4d084f6d9ad6","year":null},{"cited_arxiv_id":"2110.14168","doi":"","is_internal_anchor":true,"ref_index":5,"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","year":null}],"snapshot_sha256":"1a1ffd0dff706fbc80115a2d8f45ae0744dca908e7641f8ffa625ad4cf6fe554"},"source":{"id":"2504.11456","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-16T10:27:45.145056Z","id":"ee96a8e9-1767-4fb8-93f7-e4cf7534a95e","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"DeepMath-103K supplies 103K hard, clean math problems that let reinforcement learning reach state-of-the-art reasoning performance.","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","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."}},"verdict_id":"ee96a8e9-1767-4fb8-93f7-e4cf7534a95e"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c4930c0caa82e2cecee8c0843be4ee3bbe9439c5e705662abe2f38d3a56519a9","target":"record","created_at":"2026-05-17T23:38:48Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"07c276a3630efe36049efdcd5a7b0393561f3daaeb3ad81157d140ec6a7b5b34","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-15T17:59:51Z","title_canon_sha256":"53e33e5e58a1bb68b4dc91d0f24d06d1a094d9a28713b715d476eeb3c32a2cf6"},"schema_version":"1.0","source":{"id":"2504.11456","kind":"arxiv","version":2}},"canonical_sha256":"f8d889465f9ea6c9567ade616662015fcf8899325d78cadaaa64e4f713039c33","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f8d889465f9ea6c9567ade616662015fcf8899325d78cadaaa64e4f713039c33","first_computed_at":"2026-05-17T23:38:48.188069Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:48.188069Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"J/wbu5N2RpLYetNbRIGl6L+ZLKY5+puxlR0eH43P/C+w5KAmoibdsaT1vocthz4ttHI7K+E8EyMZuMGbMubDBA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:48.188571Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.11456","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c4930c0caa82e2cecee8c0843be4ee3bbe9439c5e705662abe2f38d3a56519a9","sha256:03dfb9e7aa746cf4d0b5cba7e1bf77889752bb79de14b0c9c35b9ff99d79644f"],"state_sha256":"2cf2e5aa1770a07e0266d5462999378eb267955f5b6aba022deea789231e27c4"}