{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XTA4ZPS74BUTUUQJP5B4LD476M","short_pith_number":"pith:XTA4ZPS7","schema_version":"1.0","canonical_sha256":"bcc1ccbe5fe0693a52097f43c58f9ff30ce981ac24aeb8add96286f184502e70","source":{"kind":"arxiv","id":"2502.19058","version":1},"attestation_state":"computed","paper":{"title":"MathClean: A Benchmark for Synthetic Mathematical Data Cleaning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Cui, Hao Liang, Meiyi Qiang, Wentao Zhang, Yongzhen Guo, Yuying Li, Zefeng He, Zhengzhou Zhu","submitted_at":"2025-02-26T11:17:50Z","abstract_excerpt":"With the rapid development of large language models (LLMs), the quality of training data has become crucial. Among the various types of training data, mathematical data plays a key role in enabling LLMs to acquire strong reasoning abilities. While high-quality open-source data is important, it is often insufficient for pre-training, necessitating the addition of synthetic math problems. However, synthetic math questions and answers can introduce inaccuracies, which may degrade both the training data and web data. Therefore, an effective method for cleaning synthetic math data is essential. In "},"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":"2502.19058","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-26T11:17:50Z","cross_cats_sorted":[],"title_canon_sha256":"d020951abcffefb647de866619d70aeccf745dbaa3459824d7c8d3e5fc2beb69","abstract_canon_sha256":"d828a327aed5f0a0f9f3eb86ddf5e77c6b8a211228950c414867b8651c4c199b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:20:26.075470Z","signature_b64":"rD2VL2afhKxKqk0DWRkH1LBOiWSOb0TlX5y+lQZoIg3u5ng/LCDeI4V5zK5wv04tqGmvT4zrIbMpk7g/QCdVAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bcc1ccbe5fe0693a52097f43c58f9ff30ce981ac24aeb8add96286f184502e70","last_reissued_at":"2026-07-05T10:20:26.074932Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:20:26.074932Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MathClean: A Benchmark for Synthetic Mathematical Data Cleaning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Cui, Hao Liang, Meiyi Qiang, Wentao Zhang, Yongzhen Guo, Yuying Li, Zefeng He, Zhengzhou Zhu","submitted_at":"2025-02-26T11:17:50Z","abstract_excerpt":"With the rapid development of large language models (LLMs), the quality of training data has become crucial. Among the various types of training data, mathematical data plays a key role in enabling LLMs to acquire strong reasoning abilities. While high-quality open-source data is important, it is often insufficient for pre-training, necessitating the addition of synthetic math problems. However, synthetic math questions and answers can introduce inaccuracies, which may degrade both the training data and web data. Therefore, an effective method for cleaning synthetic math data is essential. In "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.19058","kind":"arxiv","version":1},"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/2502.19058/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":"2502.19058","created_at":"2026-07-05T10:20:26.074998+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.19058v1","created_at":"2026-07-05T10:20:26.074998+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.19058","created_at":"2026-07-05T10:20:26.074998+00:00"},{"alias_kind":"pith_short_12","alias_value":"XTA4ZPS74BUT","created_at":"2026-07-05T10:20:26.074998+00:00"},{"alias_kind":"pith_short_16","alias_value":"XTA4ZPS74BUTUUQJ","created_at":"2026-07-05T10:20:26.074998+00:00"},{"alias_kind":"pith_short_8","alias_value":"XTA4ZPS7","created_at":"2026-07-05T10:20:26.074998+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.04707","citing_title":"OpenWorldLib: A Unified Codebase and Definition of Advanced World Models","ref_index":76,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M","json":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M.json","graph_json":"https://pith.science/api/pith-number/XTA4ZPS74BUTUUQJP5B4LD476M/graph.json","events_json":"https://pith.science/api/pith-number/XTA4ZPS74BUTUUQJP5B4LD476M/events.json","paper":"https://pith.science/paper/XTA4ZPS7"},"agent_actions":{"view_html":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M","download_json":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M.json","view_paper":"https://pith.science/paper/XTA4ZPS7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.19058&json=true","fetch_graph":"https://pith.science/api/pith-number/XTA4ZPS74BUTUUQJP5B4LD476M/graph.json","fetch_events":"https://pith.science/api/pith-number/XTA4ZPS74BUTUUQJP5B4LD476M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M/action/storage_attestation","attest_author":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M/action/author_attestation","sign_citation":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M/action/citation_signature","submit_replication":"https://pith.science/pith/XTA4ZPS74BUTUUQJP5B4LD476M/action/replication_record"}},"created_at":"2026-07-05T10:20:26.074998+00:00","updated_at":"2026-07-05T10:20:26.074998+00:00"}