{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CY7RNBPLI2HIW63FEAK4SRKGSB","short_pith_number":"pith:CY7RNBPL","schema_version":"1.0","canonical_sha256":"163f1685eb468e8b7b652015c945469065bf24c0f2728a40f351acd80270589f","source":{"kind":"arxiv","id":"2504.20930","version":2},"attestation_state":"computed","paper":{"title":"ChestX-Reasoner: Advancing Radiology Foundation Models with Reasoning through Step-by-Step Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chaoyi Wu, Cheng Liang, Weidi Xie, Yanfeng Wang, Ya Zhang, Ziqing Fan","submitted_at":"2025-04-29T16:48:23Z","abstract_excerpt":"Recent advances in reasoning-enhanced large language models (LLMs) and multimodal LLMs (MLLMs) have significantly improved performance in complex tasks, yet medical AI models often overlook the structured reasoning processes inherent in clinical practice. In this work, we present ChestX-Reasoner, a radiology diagnosis MLLM designed to leverage process supervision mined directly from clinical reports, reflecting the step-by-step reasoning followed by radiologists. We construct a large dataset by extracting and refining reasoning chains from routine radiology reports. Our two-stage training fram"},"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":"2504.20930","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-04-29T16:48:23Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"61c2e91bb34a19985f779ba97117d0cae805da8ccae2d9a0784e20ea30d8766f","abstract_canon_sha256":"ded5294965c09b5541122205b523c9090aba9876c3edd81400dcaf6fe8f85627"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:31.178973Z","signature_b64":"C8a7YZqROx0lrZjegfgwf8ECIdI+3nOhmHpHuq9iuv2PBSULINgOCyJ5pg2pUCi6hUbz2rBbpb2NH2krWzCbAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"163f1685eb468e8b7b652015c945469065bf24c0f2728a40f351acd80270589f","last_reissued_at":"2026-07-05T11:06:31.178481Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:31.178481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ChestX-Reasoner: Advancing Radiology Foundation Models with Reasoning through Step-by-Step Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chaoyi Wu, Cheng Liang, Weidi Xie, Yanfeng Wang, Ya Zhang, Ziqing Fan","submitted_at":"2025-04-29T16:48:23Z","abstract_excerpt":"Recent advances in reasoning-enhanced large language models (LLMs) and multimodal LLMs (MLLMs) have significantly improved performance in complex tasks, yet medical AI models often overlook the structured reasoning processes inherent in clinical practice. In this work, we present ChestX-Reasoner, a radiology diagnosis MLLM designed to leverage process supervision mined directly from clinical reports, reflecting the step-by-step reasoning followed by radiologists. We construct a large dataset by extracting and refining reasoning chains from routine radiology reports. Our two-stage training fram"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.20930","kind":"arxiv","version":2},"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/2504.20930/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":"2504.20930","created_at":"2026-07-05T11:06:31.178541+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.20930v2","created_at":"2026-07-05T11:06:31.178541+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.20930","created_at":"2026-07-05T11:06:31.178541+00:00"},{"alias_kind":"pith_short_12","alias_value":"CY7RNBPLI2HI","created_at":"2026-07-05T11:06:31.178541+00:00"},{"alias_kind":"pith_short_16","alias_value":"CY7RNBPLI2HIW63F","created_at":"2026-07-05T11:06:31.178541+00:00"},{"alias_kind":"pith_short_8","alias_value":"CY7RNBPL","created_at":"2026-07-05T11:06:31.178541+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07761","citing_title":"Aligning Clinical Needs and AI Capabilities: A Survey on LLMs for Medical Reasoning","ref_index":121,"is_internal_anchor":true},{"citing_arxiv_id":"2606.06379","citing_title":"EasyLens: A Training-Free Plug-and-Play Subtle-Lesion Representation Amplifier for Medical Vision-Language Models","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20277","citing_title":"Regulating Anatomy-Aware Rewards via Trajectory-Integral Feedback for Volumetric Computed Tomography Analysis","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2509.08827","citing_title":"A Survey of Reinforcement Learning for Large Reasoning Models","ref_index":134,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00876","citing_title":"GAZE: Grounded Agentic Zero-shot Evaluation with Viewer-Level Tools and Literature Retrieval on Rare Brain MRI","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB","json":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB.json","graph_json":"https://pith.science/api/pith-number/CY7RNBPLI2HIW63FEAK4SRKGSB/graph.json","events_json":"https://pith.science/api/pith-number/CY7RNBPLI2HIW63FEAK4SRKGSB/events.json","paper":"https://pith.science/paper/CY7RNBPL"},"agent_actions":{"view_html":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB","download_json":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB.json","view_paper":"https://pith.science/paper/CY7RNBPL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.20930&json=true","fetch_graph":"https://pith.science/api/pith-number/CY7RNBPLI2HIW63FEAK4SRKGSB/graph.json","fetch_events":"https://pith.science/api/pith-number/CY7RNBPLI2HIW63FEAK4SRKGSB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB/action/storage_attestation","attest_author":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB/action/author_attestation","sign_citation":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB/action/citation_signature","submit_replication":"https://pith.science/pith/CY7RNBPLI2HIW63FEAK4SRKGSB/action/replication_record"}},"created_at":"2026-07-05T11:06:31.178541+00:00","updated_at":"2026-07-05T11:06:31.178541+00:00"}