{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2G4VOEEENSY4CLZ2Z26J4SSPX4","short_pith_number":"pith:2G4VOEEE","schema_version":"1.0","canonical_sha256":"d1b95710846cb1c12f3acebc9e4a4fbf20326bb6863f4b242710dc9bc44061be","source":{"kind":"arxiv","id":"2603.26028","version":2},"attestation_state":"computed","paper":{"title":"Learning to Trim: End-to-End Causal Graph Pruning with Dynamic Anatomical Feature Banks for Medical VQA","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Qiang Li, Weizhi Nie, Yuting Su, Zibo Xu","submitted_at":"2026-03-27T02:57:39Z","abstract_excerpt":"Medical Visual Question Answering (MedVQA) models often exhibit limited generalization due to reliance on dataset-specific correlations, such as recurring anatomical patterns or question-type regularities, rather than genuine diagnostic evidence. Existing causal approaches are typically implemented as static adjustments or post-hoc corrections. To address this issue, we propose a Learnable Causal Trimming (LCT) framework that integrates causal pruning into end-to-end optimization. We introduce a Dynamic Anatomical Feature Bank (DAFB), updated via a momentum mechanism, to capture global prototy"},"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":"2603.26028","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-27T02:57:39Z","cross_cats_sorted":[],"title_canon_sha256":"d252468170bb0ccab94b75e9ff6686fbebf77de43b19ec86c64ab4f755830680","abstract_canon_sha256":"ae72ba8b998c79141575303f73fb4355c3f1daaaf18497af1c98db01a9692d1c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:04:16.189038Z","signature_b64":"7nXPN2MJAgd7QJBP1/NW7+UGiVGuege+NP6218uuS5FeWglqU8DY5x79DR/FNvWDhQoYpPleVCJy44wPYuOWCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d1b95710846cb1c12f3acebc9e4a4fbf20326bb6863f4b242710dc9bc44061be","last_reissued_at":"2026-06-02T01:04:16.188544Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:04:16.188544Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Trim: End-to-End Causal Graph Pruning with Dynamic Anatomical Feature Banks for Medical VQA","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Qiang Li, Weizhi Nie, Yuting Su, Zibo Xu","submitted_at":"2026-03-27T02:57:39Z","abstract_excerpt":"Medical Visual Question Answering (MedVQA) models often exhibit limited generalization due to reliance on dataset-specific correlations, such as recurring anatomical patterns or question-type regularities, rather than genuine diagnostic evidence. Existing causal approaches are typically implemented as static adjustments or post-hoc corrections. To address this issue, we propose a Learnable Causal Trimming (LCT) framework that integrates causal pruning into end-to-end optimization. We introduce a Dynamic Anatomical Feature Bank (DAFB), updated via a momentum mechanism, to capture global prototy"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.26028","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/2603.26028/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":"2603.26028","created_at":"2026-06-02T01:04:16.188602+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.26028v2","created_at":"2026-06-02T01:04:16.188602+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.26028","created_at":"2026-06-02T01:04:16.188602+00:00"},{"alias_kind":"pith_short_12","alias_value":"2G4VOEEENSY4","created_at":"2026-06-02T01:04:16.188602+00:00"},{"alias_kind":"pith_short_16","alias_value":"2G4VOEEENSY4CLZ2","created_at":"2026-06-02T01:04:16.188602+00:00"},{"alias_kind":"pith_short_8","alias_value":"2G4VOEEE","created_at":"2026-06-02T01:04:16.188602+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/2G4VOEEENSY4CLZ2Z26J4SSPX4","json":"https://pith.science/pith/2G4VOEEENSY4CLZ2Z26J4SSPX4.json","graph_json":"https://pith.science/api/pith-number/2G4VOEEENSY4CLZ2Z26J4SSPX4/graph.json","events_json":"https://pith.science/api/pith-number/2G4VOEEENSY4CLZ2Z26J4SSPX4/events.json","paper":"https://pith.science/paper/2G4VOEEE"},"agent_actions":{"view_html":"https://pith.science/pith/2G4VOEEENSY4CLZ2Z26J4SSPX4","download_json":"https://pith.science/pith/2G4VOEEENSY4CLZ2Z26J4SSPX4.json","view_paper":"https://pith.science/paper/2G4VOEEE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.26028&json=true","fetch_graph":"https://pith.science/api/pith-number/2G4VOEEENSY4CLZ2Z26J4SSPX4/graph.json","fetch_events":"https://pith.science/api/pith-number/2G4VOEEENSY4CLZ2Z26J4SSPX4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2G4VOEEENSY4CLZ2Z26J4SSPX4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2G4VOEEENSY4CLZ2Z26J4SSPX4/action/storage_attestation","attest_author":"https://pith.science/pith/2G4VOEEENSY4CLZ2Z26J4SSPX4/action/author_attestation","sign_citation":"https://pith.science/pith/2G4VOEEENSY4CLZ2Z26J4SSPX4/action/citation_signature","submit_replication":"https://pith.science/pith/2G4VOEEENSY4CLZ2Z26J4SSPX4/action/replication_record"}},"created_at":"2026-06-02T01:04:16.188602+00:00","updated_at":"2026-06-02T01:04:16.188602+00:00"}