{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ALDQSNIMU2JZHE2AAU2KVJWSKD","short_pith_number":"pith:ALDQSNIM","schema_version":"1.0","canonical_sha256":"02c709350ca6939393400534aaa6d250e495c0b4025a0c34d80c1c40e3a0b57d","source":{"kind":"arxiv","id":"2603.12046","version":2},"attestation_state":"computed","paper":{"title":"Dr. SHAP-AV: Decoding Relative Modality Contributions via Shapley Attribution in Audio-Visual Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.SD"],"primary_cat":"eess.AS","authors_text":"Maja Pantic, Stavros Petridis, Umberto Cappellazzo","submitted_at":"2026-03-12T15:22:27Z","abstract_excerpt":"Audio-Visual Speech Recognition (AVSR) leverages both acoustic and visual information for robust recognition under noise. However, how models balance these modalities remains unclear. We present Dr. SHAP-AV, a framework using Shapley values to analyze modality contributions in AVSR. Through experiments on six models across two benchmarks and varying SNR levels, we introduce three analyses: Global SHAP for overall modality balance, Generative SHAP for contribution dynamics during decoding, and Temporal Alignment SHAP for input-output correspondence. Our findings reveal that models shift toward "},"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":"2603.12046","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2026-03-12T15:22:27Z","cross_cats_sorted":["cs.CV","cs.SD"],"title_canon_sha256":"c1c729f0e083055fb43c04131629d94d7ee644c1a4681baa2cc658006b22378d","abstract_canon_sha256":"6e6c2c5ec56d7bd054cb38980280f487673c68d9cfaa9d9cac921b19336ca9f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:15.871470Z","signature_b64":"8i6MLYlXMLzUDOo6tRlUP9lAVbCpugVnSRdPkwGckZZRE5ixf+BAmaLS5UI7zOSyr8SPEakYrQLfvhsCicBvBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"02c709350ca6939393400534aaa6d250e495c0b4025a0c34d80c1c40e3a0b57d","last_reissued_at":"2026-06-09T01:05:15.870908Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:15.870908Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dr. SHAP-AV: Decoding Relative Modality Contributions via Shapley Attribution in Audio-Visual Speech Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.SD"],"primary_cat":"eess.AS","authors_text":"Maja Pantic, Stavros Petridis, Umberto Cappellazzo","submitted_at":"2026-03-12T15:22:27Z","abstract_excerpt":"Audio-Visual Speech Recognition (AVSR) leverages both acoustic and visual information for robust recognition under noise. However, how models balance these modalities remains unclear. We present Dr. SHAP-AV, a framework using Shapley values to analyze modality contributions in AVSR. Through experiments on six models across two benchmarks and varying SNR levels, we introduce three analyses: Global SHAP for overall modality balance, Generative SHAP for contribution dynamics during decoding, and Temporal Alignment SHAP for input-output correspondence. Our findings reveal that models shift toward "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.12046","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/2603.12046/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":"2603.12046","created_at":"2026-06-09T01:05:15.870979+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.12046v2","created_at":"2026-06-09T01:05:15.870979+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.12046","created_at":"2026-06-09T01:05:15.870979+00:00"},{"alias_kind":"pith_short_12","alias_value":"ALDQSNIMU2JZ","created_at":"2026-06-09T01:05:15.870979+00:00"},{"alias_kind":"pith_short_16","alias_value":"ALDQSNIMU2JZHE2A","created_at":"2026-06-09T01:05:15.870979+00:00"},{"alias_kind":"pith_short_8","alias_value":"ALDQSNIM","created_at":"2026-06-09T01:05:15.870979+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/ALDQSNIMU2JZHE2AAU2KVJWSKD","json":"https://pith.science/pith/ALDQSNIMU2JZHE2AAU2KVJWSKD.json","graph_json":"https://pith.science/api/pith-number/ALDQSNIMU2JZHE2AAU2KVJWSKD/graph.json","events_json":"https://pith.science/api/pith-number/ALDQSNIMU2JZHE2AAU2KVJWSKD/events.json","paper":"https://pith.science/paper/ALDQSNIM"},"agent_actions":{"view_html":"https://pith.science/pith/ALDQSNIMU2JZHE2AAU2KVJWSKD","download_json":"https://pith.science/pith/ALDQSNIMU2JZHE2AAU2KVJWSKD.json","view_paper":"https://pith.science/paper/ALDQSNIM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.12046&json=true","fetch_graph":"https://pith.science/api/pith-number/ALDQSNIMU2JZHE2AAU2KVJWSKD/graph.json","fetch_events":"https://pith.science/api/pith-number/ALDQSNIMU2JZHE2AAU2KVJWSKD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ALDQSNIMU2JZHE2AAU2KVJWSKD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ALDQSNIMU2JZHE2AAU2KVJWSKD/action/storage_attestation","attest_author":"https://pith.science/pith/ALDQSNIMU2JZHE2AAU2KVJWSKD/action/author_attestation","sign_citation":"https://pith.science/pith/ALDQSNIMU2JZHE2AAU2KVJWSKD/action/citation_signature","submit_replication":"https://pith.science/pith/ALDQSNIMU2JZHE2AAU2KVJWSKD/action/replication_record"}},"created_at":"2026-06-09T01:05:15.870979+00:00","updated_at":"2026-06-09T01:05:15.870979+00:00"}