{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:H7YCQHAU7JQZ44XIG6WHEPF6IK","short_pith_number":"pith:H7YCQHAU","schema_version":"1.0","canonical_sha256":"3ff0281c14fa619e72e837ac723cbe428d3aaecda0c012f56345294aada81de0","source":{"kind":"arxiv","id":"2501.14002","version":3},"attestation_state":"computed","paper":{"title":"Advancing Mathematical Reasoning in Language Models: The Impact of Problem-Solving Data, Data Synthesis Methods, and Training Stages","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Mi Tian, Qing Tong, Tianqiao Liu, Weiqi Luo, Zitao Liu, Zui Chen","submitted_at":"2025-01-23T12:14:57Z","abstract_excerpt":"Mathematical reasoning remains a challenging area for large language models (LLMs), prompting the development of math-specific LLMs such as LLEMMA, DeepSeekMath, and Qwen2-Math, among others. These models typically follow a two-stage training paradigm: pre-training with math-related corpora and post-training with problem datasets for supervised fine-tuning (SFT). Despite these efforts, the improvements in mathematical reasoning achieved through continued pre-training (CPT) are often less significant compared to those obtained via SFT. This study addresses this discrepancy by exploring alternat"},"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":"2501.14002","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-23T12:14:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9fcfa235bcd95ec1a8df3a0f4d7630bb2a29fd4e5f8c6eed43947ef7a7d49b27","abstract_canon_sha256":"de9f6a9db24ab024bf8f9d105274ed769a7981cf0cfe44840cd4f50d5c028048"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:37:43.450988Z","signature_b64":"inxNf7/lcb2eLWozz0RhdEwmKgktJuD2erAuSFb7H2PCrRLmtU2qosrFpUecUlnOYYi/McdYpLUb+eUAm6h7Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ff0281c14fa619e72e837ac723cbe428d3aaecda0c012f56345294aada81de0","last_reissued_at":"2026-07-05T10:37:43.450343Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:37:43.450343Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Advancing Mathematical Reasoning in Language Models: The Impact of Problem-Solving Data, Data Synthesis Methods, and Training Stages","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Mi Tian, Qing Tong, Tianqiao Liu, Weiqi Luo, Zitao Liu, Zui Chen","submitted_at":"2025-01-23T12:14:57Z","abstract_excerpt":"Mathematical reasoning remains a challenging area for large language models (LLMs), prompting the development of math-specific LLMs such as LLEMMA, DeepSeekMath, and Qwen2-Math, among others. These models typically follow a two-stage training paradigm: pre-training with math-related corpora and post-training with problem datasets for supervised fine-tuning (SFT). Despite these efforts, the improvements in mathematical reasoning achieved through continued pre-training (CPT) are often less significant compared to those obtained via SFT. This study addresses this discrepancy by exploring alternat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.14002","kind":"arxiv","version":3},"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/2501.14002/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":"2501.14002","created_at":"2026-07-05T10:37:43.450417+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.14002v3","created_at":"2026-07-05T10:37:43.450417+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.14002","created_at":"2026-07-05T10:37:43.450417+00:00"},{"alias_kind":"pith_short_12","alias_value":"H7YCQHAU7JQZ","created_at":"2026-07-05T10:37:43.450417+00:00"},{"alias_kind":"pith_short_16","alias_value":"H7YCQHAU7JQZ44XI","created_at":"2026-07-05T10:37:43.450417+00:00"},{"alias_kind":"pith_short_8","alias_value":"H7YCQHAU","created_at":"2026-07-05T10:37:43.450417+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2503.09567","citing_title":"Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models","ref_index":118,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK","json":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK.json","graph_json":"https://pith.science/api/pith-number/H7YCQHAU7JQZ44XIG6WHEPF6IK/graph.json","events_json":"https://pith.science/api/pith-number/H7YCQHAU7JQZ44XIG6WHEPF6IK/events.json","paper":"https://pith.science/paper/H7YCQHAU"},"agent_actions":{"view_html":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK","download_json":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK.json","view_paper":"https://pith.science/paper/H7YCQHAU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.14002&json=true","fetch_graph":"https://pith.science/api/pith-number/H7YCQHAU7JQZ44XIG6WHEPF6IK/graph.json","fetch_events":"https://pith.science/api/pith-number/H7YCQHAU7JQZ44XIG6WHEPF6IK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK/action/storage_attestation","attest_author":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK/action/author_attestation","sign_citation":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK/action/citation_signature","submit_replication":"https://pith.science/pith/H7YCQHAU7JQZ44XIG6WHEPF6IK/action/replication_record"}},"created_at":"2026-07-05T10:37:43.450417+00:00","updated_at":"2026-07-05T10:37:43.450417+00:00"}