{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5GHEUMWUTW4DOIO3IPCVY3QC4L","short_pith_number":"pith:5GHEUMWU","schema_version":"1.0","canonical_sha256":"e98e4a32d49db83721db43c55c6e02e2d75b7ba6a994e43692f211ebb0f01021","source":{"kind":"arxiv","id":"2605.28422","version":1},"attestation_state":"computed","paper":{"title":"VITAL: Visual-Semantic Dual Supervision for Enhanced and Interpretable Latent Reasoning in Medical MLLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Haoran Sun, Jianwei Yin, Jintao Chen, Qiaoru Li, Shaotian Liang, Yankai Jiang, Yuxiang Cai","submitted_at":"2026-05-27T12:53:13Z","abstract_excerpt":"Latent reasoning enables reasoning over continuous hidden states rather than explicit tokens, avoiding the language bottleneck and inference overhead of chain-of-thought for medical VQA. However, existing methods suffer from modality collapse, insufficient visual supervision, and train-inference mismatch. Moreover, their opaque latent states offer no interpretability, which is critical in clinical applications. We propose VITAL, a latent-space reasoning framework for medical MLLMs with visual-semantic dual supervision: an auxiliary text decoder reconstructs reasoning chains from latent states,"},"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":"2605.28422","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-27T12:53:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3001873c78fdc9708ac57e508a8b3e0bd4a7f5cba8d872eae846f0cf4174f923","abstract_canon_sha256":"61dc52610a4d6989e64bd285155377ca0187a0bab39215408d8e50fc3bf8310f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:05:17.733437Z","signature_b64":"HH6Iq4Qh0qk4cm3bTTbKyps0jIcBdXM1wOBLfhN8XdGDFLfCJRzHq4tN0UQRWsG2iGcc2IXI6yIUfTEGjpOrAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e98e4a32d49db83721db43c55c6e02e2d75b7ba6a994e43692f211ebb0f01021","last_reissued_at":"2026-05-28T01:05:17.733023Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:05:17.733023Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VITAL: Visual-Semantic Dual Supervision for Enhanced and Interpretable Latent Reasoning in Medical MLLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Haoran Sun, Jianwei Yin, Jintao Chen, Qiaoru Li, Shaotian Liang, Yankai Jiang, Yuxiang Cai","submitted_at":"2026-05-27T12:53:13Z","abstract_excerpt":"Latent reasoning enables reasoning over continuous hidden states rather than explicit tokens, avoiding the language bottleneck and inference overhead of chain-of-thought for medical VQA. However, existing methods suffer from modality collapse, insufficient visual supervision, and train-inference mismatch. Moreover, their opaque latent states offer no interpretability, which is critical in clinical applications. We propose VITAL, a latent-space reasoning framework for medical MLLMs with visual-semantic dual supervision: an auxiliary text decoder reconstructs reasoning chains from latent states,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28422","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/2605.28422/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":"2605.28422","created_at":"2026-05-28T01:05:17.733080+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.28422v1","created_at":"2026-05-28T01:05:17.733080+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28422","created_at":"2026-05-28T01:05:17.733080+00:00"},{"alias_kind":"pith_short_12","alias_value":"5GHEUMWUTW4D","created_at":"2026-05-28T01:05:17.733080+00:00"},{"alias_kind":"pith_short_16","alias_value":"5GHEUMWUTW4DOIO3","created_at":"2026-05-28T01:05:17.733080+00:00"},{"alias_kind":"pith_short_8","alias_value":"5GHEUMWU","created_at":"2026-05-28T01:05:17.733080+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L","json":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L.json","graph_json":"https://pith.science/api/pith-number/5GHEUMWUTW4DOIO3IPCVY3QC4L/graph.json","events_json":"https://pith.science/api/pith-number/5GHEUMWUTW4DOIO3IPCVY3QC4L/events.json","paper":"https://pith.science/paper/5GHEUMWU"},"agent_actions":{"view_html":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L","download_json":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L.json","view_paper":"https://pith.science/paper/5GHEUMWU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.28422&json=true","fetch_graph":"https://pith.science/api/pith-number/5GHEUMWUTW4DOIO3IPCVY3QC4L/graph.json","fetch_events":"https://pith.science/api/pith-number/5GHEUMWUTW4DOIO3IPCVY3QC4L/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L/action/storage_attestation","attest_author":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L/action/author_attestation","sign_citation":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L/action/citation_signature","submit_replication":"https://pith.science/pith/5GHEUMWUTW4DOIO3IPCVY3QC4L/action/replication_record"}},"created_at":"2026-05-28T01:05:17.733080+00:00","updated_at":"2026-05-28T01:05:17.733080+00:00"}