{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3AC6CYZD7RXSNRDOFWIPSHL755","short_pith_number":"pith:3AC6CYZD","schema_version":"1.0","canonical_sha256":"d805e16323fc6f26c46e2d90f91d7fef76703566ec20b4db17c5cdc21ac5c42e","source":{"kind":"arxiv","id":"2501.11284","version":1},"attestation_state":"computed","paper":{"title":"RedStar: Does Scaling Long-CoT Data Unlock Better Slow-Reasoning Systems?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Boyuan Chen, Da Zheng, Debing Zhang, Haotian Xu, Jiaming Ji, Muhan Zhang, Shijia Kang, Weinong Wang, Xing Wu, Yaodong Yang, Yi Hu, Yingying Zhang, Zhijiang Guo, Zhongzhi Li","submitted_at":"2025-01-20T05:44:01Z","abstract_excerpt":"Can scaling transform reasoning? In this work, we explore the untapped potential of scaling Long Chain-of-Thought (Long-CoT) data to 1000k samples, pioneering the development of a slow-thinking model, RedStar. Through extensive experiments with various LLMs and different sizes, we uncover the ingredients for specialization and scale for Long-CoT training. Surprisingly, even smaller models show significant performance gains with limited data, revealing the sample efficiency of Long-CoT and the critical role of sample difficulty in the learning process. Our findings demonstrate that Long-CoT rea"},"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":"2501.11284","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-20T05:44:01Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"b569c4106b5ac3521fcd021ef3363ceed87d9ba895a2da1e0db0fb27c01a5355","abstract_canon_sha256":"c7f90bfc3682a28bcd086e8793f70a9eea7e4a7f8a2bd8384e8b168e13ad6620"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:03:06.288483Z","signature_b64":"GLYHIFwp5Rkgaf6sB3FxTrvLAfPhXleDMt1+5oRrWhEs1J7D9t+qg5YXxmHjwNua8/QRXRZX6o0DDzrPZlstAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d805e16323fc6f26c46e2d90f91d7fef76703566ec20b4db17c5cdc21ac5c42e","last_reissued_at":"2026-07-05T10:03:06.288056Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:03:06.288056Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RedStar: Does Scaling Long-CoT Data Unlock Better Slow-Reasoning Systems?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Boyuan Chen, Da Zheng, Debing Zhang, Haotian Xu, Jiaming Ji, Muhan Zhang, Shijia Kang, Weinong Wang, Xing Wu, Yaodong Yang, Yi Hu, Yingying Zhang, Zhijiang Guo, Zhongzhi Li","submitted_at":"2025-01-20T05:44:01Z","abstract_excerpt":"Can scaling transform reasoning? In this work, we explore the untapped potential of scaling Long Chain-of-Thought (Long-CoT) data to 1000k samples, pioneering the development of a slow-thinking model, RedStar. Through extensive experiments with various LLMs and different sizes, we uncover the ingredients for specialization and scale for Long-CoT training. Surprisingly, even smaller models show significant performance gains with limited data, revealing the sample efficiency of Long-CoT and the critical role of sample difficulty in the learning process. Our findings demonstrate that Long-CoT rea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.11284","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/2501.11284/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":"2501.11284","created_at":"2026-07-05T10:03:06.288111+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.11284v1","created_at":"2026-07-05T10:03:06.288111+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.11284","created_at":"2026-07-05T10:03:06.288111+00:00"},{"alias_kind":"pith_short_12","alias_value":"3AC6CYZD7RXS","created_at":"2026-07-05T10:03:06.288111+00:00"},{"alias_kind":"pith_short_16","alias_value":"3AC6CYZD7RXSNRDO","created_at":"2026-07-05T10:03:06.288111+00:00"},{"alias_kind":"pith_short_8","alias_value":"3AC6CYZD","created_at":"2026-07-05T10:03:06.288111+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.02234","citing_title":"Purified OPSD: On-Policy Self-Distillation Without Losing How to Think","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09709","citing_title":"IS-CoT: Breaking the Long-form Generation Collapse via Interleaved Structural Thinking","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2503.12605","citing_title":"Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey","ref_index":234,"is_internal_anchor":false},{"citing_arxiv_id":"2502.17419","citing_title":"From System 1 to System 2: A Survey of Reasoning Large Language Models","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03677","citing_title":"Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755","json":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755.json","graph_json":"https://pith.science/api/pith-number/3AC6CYZD7RXSNRDOFWIPSHL755/graph.json","events_json":"https://pith.science/api/pith-number/3AC6CYZD7RXSNRDOFWIPSHL755/events.json","paper":"https://pith.science/paper/3AC6CYZD"},"agent_actions":{"view_html":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755","download_json":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755.json","view_paper":"https://pith.science/paper/3AC6CYZD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.11284&json=true","fetch_graph":"https://pith.science/api/pith-number/3AC6CYZD7RXSNRDOFWIPSHL755/graph.json","fetch_events":"https://pith.science/api/pith-number/3AC6CYZD7RXSNRDOFWIPSHL755/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755/action/storage_attestation","attest_author":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755/action/author_attestation","sign_citation":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755/action/citation_signature","submit_replication":"https://pith.science/pith/3AC6CYZD7RXSNRDOFWIPSHL755/action/replication_record"}},"created_at":"2026-07-05T10:03:06.288111+00:00","updated_at":"2026-07-05T10:03:06.288111+00:00"}