{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IJGVA7UPC647PFQ2UY6YU3XTWT","short_pith_number":"pith:IJGVA7UP","schema_version":"1.0","canonical_sha256":"424d507e8f17b9f7961aa63d8a6ef3b4c81c0b0a8102dbb36b279a5a35450ea9","source":{"kind":"arxiv","id":"2506.17903","version":1},"attestation_state":"computed","paper":{"title":"Cause-Effect Driven Optimization for Robust Medical Visual Question Answering with Language Biases","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bingzhi Chen, Guangming Lu, Huanjia Zhu, Xiaozhao Fang, Yishu Liu","submitted_at":"2025-06-22T05:20:34Z","abstract_excerpt":"Existing Medical Visual Question Answering (Med-VQA) models often suffer from language biases, where spurious correlations between question types and answer categories are inadvertently established. To address these issues, we propose a novel Cause-Effect Driven Optimization framework called CEDO, that incorporates three well-established mechanisms, i.e., Modality-driven Heterogeneous Optimization (MHO), Gradient-guided Modality Synergy (GMS), and Distribution-adapted Loss Rescaling (DLR), for comprehensively mitigating language biases from both causal and effectual perspectives. Specifically,"},"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":"2506.17903","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-22T05:20:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6d5b6e03c44413676e68e6cf212cf2a6b962b3b56bd9a03d4ca586a82cffa8ea","abstract_canon_sha256":"6388b16ad5ac42b4902ebf1e687e734f8a09a4b53018a15c02e9f1a446d6f6c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:25.217809Z","signature_b64":"CMNsXuKKAIIdzKKxp7QTMfINyZBxvTkzAbM1T6dCffDGfA3W7Dh3xGgxKA7a2Wnr8wVcXHV3oNMACnTux9I8Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"424d507e8f17b9f7961aa63d8a6ef3b4c81c0b0a8102dbb36b279a5a35450ea9","last_reissued_at":"2026-07-05T11:25:25.217359Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:25.217359Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cause-Effect Driven Optimization for Robust Medical Visual Question Answering with Language Biases","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bingzhi Chen, Guangming Lu, Huanjia Zhu, Xiaozhao Fang, Yishu Liu","submitted_at":"2025-06-22T05:20:34Z","abstract_excerpt":"Existing Medical Visual Question Answering (Med-VQA) models often suffer from language biases, where spurious correlations between question types and answer categories are inadvertently established. To address these issues, we propose a novel Cause-Effect Driven Optimization framework called CEDO, that incorporates three well-established mechanisms, i.e., Modality-driven Heterogeneous Optimization (MHO), Gradient-guided Modality Synergy (GMS), and Distribution-adapted Loss Rescaling (DLR), for comprehensively mitigating language biases from both causal and effectual perspectives. Specifically,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17903","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/2506.17903/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":"2506.17903","created_at":"2026-07-05T11:25:25.217417+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.17903v1","created_at":"2026-07-05T11:25:25.217417+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17903","created_at":"2026-07-05T11:25:25.217417+00:00"},{"alias_kind":"pith_short_12","alias_value":"IJGVA7UPC647","created_at":"2026-07-05T11:25:25.217417+00:00"},{"alias_kind":"pith_short_16","alias_value":"IJGVA7UPC647PFQ2","created_at":"2026-07-05T11:25:25.217417+00:00"},{"alias_kind":"pith_short_8","alias_value":"IJGVA7UP","created_at":"2026-07-05T11:25:25.217417+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/IJGVA7UPC647PFQ2UY6YU3XTWT","json":"https://pith.science/pith/IJGVA7UPC647PFQ2UY6YU3XTWT.json","graph_json":"https://pith.science/api/pith-number/IJGVA7UPC647PFQ2UY6YU3XTWT/graph.json","events_json":"https://pith.science/api/pith-number/IJGVA7UPC647PFQ2UY6YU3XTWT/events.json","paper":"https://pith.science/paper/IJGVA7UP"},"agent_actions":{"view_html":"https://pith.science/pith/IJGVA7UPC647PFQ2UY6YU3XTWT","download_json":"https://pith.science/pith/IJGVA7UPC647PFQ2UY6YU3XTWT.json","view_paper":"https://pith.science/paper/IJGVA7UP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.17903&json=true","fetch_graph":"https://pith.science/api/pith-number/IJGVA7UPC647PFQ2UY6YU3XTWT/graph.json","fetch_events":"https://pith.science/api/pith-number/IJGVA7UPC647PFQ2UY6YU3XTWT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IJGVA7UPC647PFQ2UY6YU3XTWT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IJGVA7UPC647PFQ2UY6YU3XTWT/action/storage_attestation","attest_author":"https://pith.science/pith/IJGVA7UPC647PFQ2UY6YU3XTWT/action/author_attestation","sign_citation":"https://pith.science/pith/IJGVA7UPC647PFQ2UY6YU3XTWT/action/citation_signature","submit_replication":"https://pith.science/pith/IJGVA7UPC647PFQ2UY6YU3XTWT/action/replication_record"}},"created_at":"2026-07-05T11:25:25.217417+00:00","updated_at":"2026-07-05T11:25:25.217417+00:00"}