{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2ME5JMIUGZ5SEWBT5QZU365EQN","short_pith_number":"pith:2ME5JMIU","schema_version":"1.0","canonical_sha256":"d309d4b114367b225833ec334dfba483484ea038e2e2ec972db72a40af021ec4","source":{"kind":"arxiv","id":"2509.09397","version":1},"attestation_state":"computed","paper":{"title":"Decoupling Clinical and Class-Agnostic Features for Reliable Few-Shot Adaptation under Shift","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dwarikanath Mahapatra, Mohammad Yaqub, Raza Imam, Umaima Rahman","submitted_at":"2025-09-11T12:26:57Z","abstract_excerpt":"Medical vision-language models (VLMs) offer promise for clinical decision support, yet their reliability under distribution shifts remains a major concern for safe deployment. These models often learn task-agnostic correlations due to variability in imaging protocols and free-text reports, limiting their generalizability and increasing the risk of failure in real-world settings. We propose DRiFt, a structured feature decoupling framework that explicitly separates clinically relevant signals from task-agnostic noise using parameter-efficient tuning (LoRA) and learnable prompt tokens. To enhance"},"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":"2509.09397","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-09-11T12:26:57Z","cross_cats_sorted":[],"title_canon_sha256":"181f163f3e7a0c3d6b5e852de423f30c7858508dced45a75c18834c8e88d4f16","abstract_canon_sha256":"bb0f4bd6c0662845bb538f561a5c42333d57e72736429727aca711922cd94159"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:09:35.451184Z","signature_b64":"Jsmg5jtHbSWMaCOe+sTMD3wNt34B7NA3yfCcfQVQO9iae3A9Qtsrb5OsyKhHQitd1g5e9oNGMIC5pFsifk4/Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d309d4b114367b225833ec334dfba483484ea038e2e2ec972db72a40af021ec4","last_reissued_at":"2026-07-05T12:09:35.450624Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:09:35.450624Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Decoupling Clinical and Class-Agnostic Features for Reliable Few-Shot Adaptation under Shift","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dwarikanath Mahapatra, Mohammad Yaqub, Raza Imam, Umaima Rahman","submitted_at":"2025-09-11T12:26:57Z","abstract_excerpt":"Medical vision-language models (VLMs) offer promise for clinical decision support, yet their reliability under distribution shifts remains a major concern for safe deployment. These models often learn task-agnostic correlations due to variability in imaging protocols and free-text reports, limiting their generalizability and increasing the risk of failure in real-world settings. We propose DRiFt, a structured feature decoupling framework that explicitly separates clinically relevant signals from task-agnostic noise using parameter-efficient tuning (LoRA) and learnable prompt tokens. To enhance"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.09397","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/2509.09397/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":"2509.09397","created_at":"2026-07-05T12:09:35.450688+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.09397v1","created_at":"2026-07-05T12:09:35.450688+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.09397","created_at":"2026-07-05T12:09:35.450688+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ME5JMIUGZ5S","created_at":"2026-07-05T12:09:35.450688+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ME5JMIUGZ5SEWBT","created_at":"2026-07-05T12:09:35.450688+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ME5JMIU","created_at":"2026-07-05T12:09:35.450688+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/2ME5JMIUGZ5SEWBT5QZU365EQN","json":"https://pith.science/pith/2ME5JMIUGZ5SEWBT5QZU365EQN.json","graph_json":"https://pith.science/api/pith-number/2ME5JMIUGZ5SEWBT5QZU365EQN/graph.json","events_json":"https://pith.science/api/pith-number/2ME5JMIUGZ5SEWBT5QZU365EQN/events.json","paper":"https://pith.science/paper/2ME5JMIU"},"agent_actions":{"view_html":"https://pith.science/pith/2ME5JMIUGZ5SEWBT5QZU365EQN","download_json":"https://pith.science/pith/2ME5JMIUGZ5SEWBT5QZU365EQN.json","view_paper":"https://pith.science/paper/2ME5JMIU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.09397&json=true","fetch_graph":"https://pith.science/api/pith-number/2ME5JMIUGZ5SEWBT5QZU365EQN/graph.json","fetch_events":"https://pith.science/api/pith-number/2ME5JMIUGZ5SEWBT5QZU365EQN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ME5JMIUGZ5SEWBT5QZU365EQN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ME5JMIUGZ5SEWBT5QZU365EQN/action/storage_attestation","attest_author":"https://pith.science/pith/2ME5JMIUGZ5SEWBT5QZU365EQN/action/author_attestation","sign_citation":"https://pith.science/pith/2ME5JMIUGZ5SEWBT5QZU365EQN/action/citation_signature","submit_replication":"https://pith.science/pith/2ME5JMIUGZ5SEWBT5QZU365EQN/action/replication_record"}},"created_at":"2026-07-05T12:09:35.450688+00:00","updated_at":"2026-07-05T12:09:35.450688+00:00"}