{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CREQPQGX2AJK5V4AE4VFPCXDLN","short_pith_number":"pith:CREQPQGX","schema_version":"1.0","canonical_sha256":"144907c0d7d012aed780272a578ae35b49a1486daea2da07c333ba642e6112b9","source":{"kind":"arxiv","id":"2509.06079","version":1},"attestation_state":"computed","paper":{"title":"Multimodal Reasoning for Science: Technical Report and 1st Place Solution to the ICML 2025 SeePhys Challenge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Bin Dong, Bohan Zeng, Hao Liang, Junbo Niu, Ruitao Wu, Wentao Zhang","submitted_at":"2025-09-07T14:47:32Z","abstract_excerpt":"Multimodal reasoning remains a fundamental challenge in artificial intelligence. Despite substantial advances in text-based reasoning, even state-of-the-art models such as GPT-o3 struggle to maintain strong performance in multimodal scenarios. To address this gap, we introduce a caption-assisted reasoning framework that effectively bridges visual and textual modalities. Our approach achieved 1st place in the ICML 2025 AI for Math Workshop \\& Challenge 2: SeePhys, highlighting its effectiveness and robustness. Furthermore, we validate its generalization on the MathVerse benchmark for geometric "},"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":"2509.06079","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-07T14:47:32Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"a6582e40a423f8f223a60d01c99b27462e266a3836faa34e9aab57c9f3a2d23e","abstract_canon_sha256":"9f3bf13b53a4c50b8670738bbe28c2a7158be2f52ea12cecb9ae23d7067ae8bb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:29.635300Z","signature_b64":"7AdhHbrx+t9wTQ4svCl4cEeAh2RFq4rdGPqEYRxOzyWHJ/0EMOz8XNCbh8q2BqdD44RtmUlth991cxgkUg+uAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"144907c0d7d012aed780272a578ae35b49a1486daea2da07c333ba642e6112b9","last_reissued_at":"2026-07-05T12:06:29.634744Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:29.634744Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multimodal Reasoning for Science: Technical Report and 1st Place Solution to the ICML 2025 SeePhys Challenge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Bin Dong, Bohan Zeng, Hao Liang, Junbo Niu, Ruitao Wu, Wentao Zhang","submitted_at":"2025-09-07T14:47:32Z","abstract_excerpt":"Multimodal reasoning remains a fundamental challenge in artificial intelligence. Despite substantial advances in text-based reasoning, even state-of-the-art models such as GPT-o3 struggle to maintain strong performance in multimodal scenarios. To address this gap, we introduce a caption-assisted reasoning framework that effectively bridges visual and textual modalities. Our approach achieved 1st place in the ICML 2025 AI for Math Workshop \\& Challenge 2: SeePhys, highlighting its effectiveness and robustness. Furthermore, we validate its generalization on the MathVerse benchmark for geometric "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.06079","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/2509.06079/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":"2509.06079","created_at":"2026-07-05T12:06:29.634823+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.06079v1","created_at":"2026-07-05T12:06:29.634823+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.06079","created_at":"2026-07-05T12:06:29.634823+00:00"},{"alias_kind":"pith_short_12","alias_value":"CREQPQGX2AJK","created_at":"2026-07-05T12:06:29.634823+00:00"},{"alias_kind":"pith_short_16","alias_value":"CREQPQGX2AJK5V4A","created_at":"2026-07-05T12:06:29.634823+00:00"},{"alias_kind":"pith_short_8","alias_value":"CREQPQGX","created_at":"2026-07-05T12:06:29.634823+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.14159","citing_title":"MVI-Bench: A Comprehensive Benchmark for Evaluating Robustness to Misleading Visual Inputs in LVLMs","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2510.04978","citing_title":"Aligning Perception, Reasoning, Modeling and Interaction: A Survey on Physical AI","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04707","citing_title":"OpenWorldLib: A Unified Codebase and Definition of Advanced World Models","ref_index":77,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN","json":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN.json","graph_json":"https://pith.science/api/pith-number/CREQPQGX2AJK5V4AE4VFPCXDLN/graph.json","events_json":"https://pith.science/api/pith-number/CREQPQGX2AJK5V4AE4VFPCXDLN/events.json","paper":"https://pith.science/paper/CREQPQGX"},"agent_actions":{"view_html":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN","download_json":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN.json","view_paper":"https://pith.science/paper/CREQPQGX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.06079&json=true","fetch_graph":"https://pith.science/api/pith-number/CREQPQGX2AJK5V4AE4VFPCXDLN/graph.json","fetch_events":"https://pith.science/api/pith-number/CREQPQGX2AJK5V4AE4VFPCXDLN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN/action/storage_attestation","attest_author":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN/action/author_attestation","sign_citation":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN/action/citation_signature","submit_replication":"https://pith.science/pith/CREQPQGX2AJK5V4AE4VFPCXDLN/action/replication_record"}},"created_at":"2026-07-05T12:06:29.634823+00:00","updated_at":"2026-07-05T12:06:29.634823+00:00"}