{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VQRT7L3K7UVBGW3BTZXI6UDSAB","short_pith_number":"pith:VQRT7L3K","schema_version":"1.0","canonical_sha256":"ac233faf6afd2a135b619e6e8f5072005a768c4a9e253c626e9c3e684e33c6d2","source":{"kind":"arxiv","id":"2607.09884","version":1},"attestation_state":"computed","paper":{"title":"ShapKO: Shapley-Adaptive Modality Knockout for Robust Multimodal Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Fengbei Liu, Mert R. Sabuncu, Minh Nguyen, Nusrat Binta Nizam, Ruining Deng, Sunwoo Kwak","submitted_at":"2026-07-10T18:16:20Z","abstract_excerpt":"Multimodal medical models often degrade when inputs are missing, a common scenario in real-world clinical workflows. Separately, even when all modalities are present, modality dominance is observed during training, where optimization over-relies on a highly predictive modality and undertrains complementary sources, resulting in poor robustness under partial availability. While training-time modality knockout improves missing-modality robustness, existing approaches use static masking rates that cannot adapt to evolving modality utility during training. We introduce ShapKO (Shapley-Adaptive Mod"},"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":"2607.09884","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-07-10T18:16:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"deb4f8e06b594fc2830484bd89e8d7268541d82cb63b229311d703c720dca2f6","abstract_canon_sha256":"a4f9a8198e0c096471f425c158c2c0568f3ed923220682fd06f792afb724e8c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T00:18:41.422104Z","signature_b64":"YxLeGKDZtEJkAlZ6lFy6vhDm4jffdAMo7nNZkHI76zXv0NYi5J02ujB5siIINGZdOSpanJ1us2Xhu2k7Nw7PBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac233faf6afd2a135b619e6e8f5072005a768c4a9e253c626e9c3e684e33c6d2","last_reissued_at":"2026-07-14T00:18:41.421302Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T00:18:41.421302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ShapKO: Shapley-Adaptive Modality Knockout for Robust Multimodal Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Fengbei Liu, Mert R. Sabuncu, Minh Nguyen, Nusrat Binta Nizam, Ruining Deng, Sunwoo Kwak","submitted_at":"2026-07-10T18:16:20Z","abstract_excerpt":"Multimodal medical models often degrade when inputs are missing, a common scenario in real-world clinical workflows. Separately, even when all modalities are present, modality dominance is observed during training, where optimization over-relies on a highly predictive modality and undertrains complementary sources, resulting in poor robustness under partial availability. While training-time modality knockout improves missing-modality robustness, existing approaches use static masking rates that cannot adapt to evolving modality utility during training. We introduce ShapKO (Shapley-Adaptive Mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.09884","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/2607.09884/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":"2607.09884","created_at":"2026-07-14T00:18:41.421723+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.09884v1","created_at":"2026-07-14T00:18:41.421723+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.09884","created_at":"2026-07-14T00:18:41.421723+00:00"},{"alias_kind":"pith_short_12","alias_value":"VQRT7L3K7UVB","created_at":"2026-07-14T00:18:41.421723+00:00"},{"alias_kind":"pith_short_16","alias_value":"VQRT7L3K7UVBGW3B","created_at":"2026-07-14T00:18:41.421723+00:00"},{"alias_kind":"pith_short_8","alias_value":"VQRT7L3K","created_at":"2026-07-14T00:18:41.421723+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/VQRT7L3K7UVBGW3BTZXI6UDSAB","json":"https://pith.science/pith/VQRT7L3K7UVBGW3BTZXI6UDSAB.json","graph_json":"https://pith.science/api/pith-number/VQRT7L3K7UVBGW3BTZXI6UDSAB/graph.json","events_json":"https://pith.science/api/pith-number/VQRT7L3K7UVBGW3BTZXI6UDSAB/events.json","paper":"https://pith.science/paper/VQRT7L3K"},"agent_actions":{"view_html":"https://pith.science/pith/VQRT7L3K7UVBGW3BTZXI6UDSAB","download_json":"https://pith.science/pith/VQRT7L3K7UVBGW3BTZXI6UDSAB.json","view_paper":"https://pith.science/paper/VQRT7L3K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.09884&json=true","fetch_graph":"https://pith.science/api/pith-number/VQRT7L3K7UVBGW3BTZXI6UDSAB/graph.json","fetch_events":"https://pith.science/api/pith-number/VQRT7L3K7UVBGW3BTZXI6UDSAB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VQRT7L3K7UVBGW3BTZXI6UDSAB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VQRT7L3K7UVBGW3BTZXI6UDSAB/action/storage_attestation","attest_author":"https://pith.science/pith/VQRT7L3K7UVBGW3BTZXI6UDSAB/action/author_attestation","sign_citation":"https://pith.science/pith/VQRT7L3K7UVBGW3BTZXI6UDSAB/action/citation_signature","submit_replication":"https://pith.science/pith/VQRT7L3K7UVBGW3BTZXI6UDSAB/action/replication_record"}},"created_at":"2026-07-14T00:18:41.421723+00:00","updated_at":"2026-07-14T00:18:41.421723+00:00"}