{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GQQJWON52QWJO3YH2V2BX4QAOF","short_pith_number":"pith:GQQJWON5","schema_version":"1.0","canonical_sha256":"34209b39bdd42c976f07d5741bf200715693d124f5ac5d00dac4690efba5bb41","source":{"kind":"arxiv","id":"2607.16726","version":1},"attestation_state":"computed","paper":{"title":"Can Experts Adapt Without Training? On Test-Time Modality Generalization in MVLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Brejesh Lall, Darakshan Rashid, Dwarikanath Mahapatra, Mohammad Yaqub, Raza Imam, Yutong Xie","submitted_at":"2026-07-18T09:24:39Z","abstract_excerpt":"Medical vision-language models (MVLMs) promise broad zero-shot generalization, yet their reliability collapses when confronted with unseen modalities and domains, precisely where clinical robustness matters most. To address this gap, we revisit test-time modality generalization from the perspective of Mixture-of-Experts (MoE) and ask: can experts route-and-adapt without any optimization during inference? We identify a fundamental specialization-generalization dilemma at test time, where blindly aggregating modality experts dilutes modality-specific knowledge, while selecting one highly confide"},"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":"2607.16726","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-18T09:24:39Z","cross_cats_sorted":[],"title_canon_sha256":"22f9dcfc73f5c5444b583298e2b360caa45454ce2b9667a69fe26f4cb329fbb0","abstract_canon_sha256":"ea7c72c5b34ff73e1d03fee5083c45943c879d0074b71b4155017d7906793f84"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:20:23.823752Z","signature_b64":"L2b3Psl+MGXxvxJtP8zgOqbQ0srnmjIu2T6l92lCULPoAmwePh3w2dibxurlg0wa4aSHt1uZ5yosRbJ6rMISDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"34209b39bdd42c976f07d5741bf200715693d124f5ac5d00dac4690efba5bb41","last_reissued_at":"2026-07-21T01:20:23.822860Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:20:23.822860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can Experts Adapt Without Training? On Test-Time Modality Generalization in MVLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Brejesh Lall, Darakshan Rashid, Dwarikanath Mahapatra, Mohammad Yaqub, Raza Imam, Yutong Xie","submitted_at":"2026-07-18T09:24:39Z","abstract_excerpt":"Medical vision-language models (MVLMs) promise broad zero-shot generalization, yet their reliability collapses when confronted with unseen modalities and domains, precisely where clinical robustness matters most. To address this gap, we revisit test-time modality generalization from the perspective of Mixture-of-Experts (MoE) and ask: can experts route-and-adapt without any optimization during inference? We identify a fundamental specialization-generalization dilemma at test time, where blindly aggregating modality experts dilutes modality-specific knowledge, while selecting one highly confide"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.16726","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/2607.16726/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":"2607.16726","created_at":"2026-07-21T01:20:23.823324+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.16726v1","created_at":"2026-07-21T01:20:23.823324+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.16726","created_at":"2026-07-21T01:20:23.823324+00:00"},{"alias_kind":"pith_short_12","alias_value":"GQQJWON52QWJ","created_at":"2026-07-21T01:20:23.823324+00:00"},{"alias_kind":"pith_short_16","alias_value":"GQQJWON52QWJO3YH","created_at":"2026-07-21T01:20:23.823324+00:00"},{"alias_kind":"pith_short_8","alias_value":"GQQJWON5","created_at":"2026-07-21T01:20:23.823324+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/GQQJWON52QWJO3YH2V2BX4QAOF","json":"https://pith.science/pith/GQQJWON52QWJO3YH2V2BX4QAOF.json","graph_json":"https://pith.science/api/pith-number/GQQJWON52QWJO3YH2V2BX4QAOF/graph.json","events_json":"https://pith.science/api/pith-number/GQQJWON52QWJO3YH2V2BX4QAOF/events.json","paper":"https://pith.science/paper/GQQJWON5"},"agent_actions":{"view_html":"https://pith.science/pith/GQQJWON52QWJO3YH2V2BX4QAOF","download_json":"https://pith.science/pith/GQQJWON52QWJO3YH2V2BX4QAOF.json","view_paper":"https://pith.science/paper/GQQJWON5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.16726&json=true","fetch_graph":"https://pith.science/api/pith-number/GQQJWON52QWJO3YH2V2BX4QAOF/graph.json","fetch_events":"https://pith.science/api/pith-number/GQQJWON52QWJO3YH2V2BX4QAOF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GQQJWON52QWJO3YH2V2BX4QAOF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GQQJWON52QWJO3YH2V2BX4QAOF/action/storage_attestation","attest_author":"https://pith.science/pith/GQQJWON52QWJO3YH2V2BX4QAOF/action/author_attestation","sign_citation":"https://pith.science/pith/GQQJWON52QWJO3YH2V2BX4QAOF/action/citation_signature","submit_replication":"https://pith.science/pith/GQQJWON52QWJO3YH2V2BX4QAOF/action/replication_record"}},"created_at":"2026-07-21T01:20:23.823324+00:00","updated_at":"2026-07-21T01:20:23.823324+00:00"}