{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z22ZB7RMQSP22R4E3NXATKCVKO","short_pith_number":"pith:Z22ZB7RM","schema_version":"1.0","canonical_sha256":"ceb590fe2c849fad4784db6e09a855539a3c86c534351e657bd09deacd1de0d3","source":{"kind":"arxiv","id":"2503.06252","version":1},"attestation_state":"computed","paper":{"title":"Can Atomic Step Decomposition Enhance the Self-structured Reasoning of Multimodal Large Models?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hang Xu, Hanhui Li, Haoxiang Fan, Jianhua Han, Kaixin Cai, Kun Xiang, Lanqing Hong, Runhui Huang, Weiran Huang, Xiaodan Liang, Yihan Zeng, Yiyang Yin, Yu-Jie Yuan, Yunshuang Nie, Zhili Liu, Zihao Jiang","submitted_at":"2025-03-08T15:23:47Z","abstract_excerpt":"In this paper, we address the challenging task of multimodal mathematical reasoning by incorporating the ability of \"slow thinking\" into multimodal large language models (MLLMs). Our core idea is that different levels of reasoning abilities can be combined dynamically to tackle questions with different complexity. To this end, we propose a paradigm of Self-structured Chain of Thought (SCoT), which is composed of minimal semantic atomic steps. Different from existing methods that rely on structured templates or free-form paradigms, our method can not only generate cognitive CoT structures for v"},"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":"2503.06252","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-08T15:23:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d7d977d09ca4d1f736af1f1667818f919da48b0bd100cad95f084626bf7380b6","abstract_canon_sha256":"1f8372aaed8c1e091346f9b32430facfb1634a19a36067f6849f587b0a8250ad"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:47:07.885613Z","signature_b64":"p2fQRty9pR0J1ZRNUF9aX9BK9w0rghMCCqrm1waOTqmKB1Y68RxWJE0wZVumWeNAe20APwW/Y0/ulzAr2awBBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ceb590fe2c849fad4784db6e09a855539a3c86c534351e657bd09deacd1de0d3","last_reissued_at":"2026-07-05T11:47:07.885077Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:47:07.885077Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can Atomic Step Decomposition Enhance the Self-structured Reasoning of Multimodal Large Models?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hang Xu, Hanhui Li, Haoxiang Fan, Jianhua Han, Kaixin Cai, Kun Xiang, Lanqing Hong, Runhui Huang, Weiran Huang, Xiaodan Liang, Yihan Zeng, Yiyang Yin, Yu-Jie Yuan, Yunshuang Nie, Zhili Liu, Zihao Jiang","submitted_at":"2025-03-08T15:23:47Z","abstract_excerpt":"In this paper, we address the challenging task of multimodal mathematical reasoning by incorporating the ability of \"slow thinking\" into multimodal large language models (MLLMs). Our core idea is that different levels of reasoning abilities can be combined dynamically to tackle questions with different complexity. To this end, we propose a paradigm of Self-structured Chain of Thought (SCoT), which is composed of minimal semantic atomic steps. Different from existing methods that rely on structured templates or free-form paradigms, our method can not only generate cognitive CoT structures for v"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.06252","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/2503.06252/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":"2503.06252","created_at":"2026-07-05T11:47:07.885137+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.06252v1","created_at":"2026-07-05T11:47:07.885137+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.06252","created_at":"2026-07-05T11:47:07.885137+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z22ZB7RMQSP2","created_at":"2026-07-05T11:47:07.885137+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z22ZB7RMQSP22R4E","created_at":"2026-07-05T11:47:07.885137+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z22ZB7RM","created_at":"2026-07-05T11:47:07.885137+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2503.16419","citing_title":"Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models","ref_index":195,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06165","citing_title":"Post Reasoning: Improving the Performance of Non-Thinking Models at No Cost","ref_index":160,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO","json":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO.json","graph_json":"https://pith.science/api/pith-number/Z22ZB7RMQSP22R4E3NXATKCVKO/graph.json","events_json":"https://pith.science/api/pith-number/Z22ZB7RMQSP22R4E3NXATKCVKO/events.json","paper":"https://pith.science/paper/Z22ZB7RM"},"agent_actions":{"view_html":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO","download_json":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO.json","view_paper":"https://pith.science/paper/Z22ZB7RM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.06252&json=true","fetch_graph":"https://pith.science/api/pith-number/Z22ZB7RMQSP22R4E3NXATKCVKO/graph.json","fetch_events":"https://pith.science/api/pith-number/Z22ZB7RMQSP22R4E3NXATKCVKO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO/action/storage_attestation","attest_author":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO/action/author_attestation","sign_citation":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO/action/citation_signature","submit_replication":"https://pith.science/pith/Z22ZB7RMQSP22R4E3NXATKCVKO/action/replication_record"}},"created_at":"2026-07-05T11:47:07.885137+00:00","updated_at":"2026-07-05T11:47:07.885137+00:00"}