{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PXZKU2CD2KSD2X3S2LEUMDQSI4","short_pith_number":"pith:PXZKU2CD","schema_version":"1.0","canonical_sha256":"7df2aa6843d2a43d5f72d2c9460e1247079a808aa589989a214d173fea0e2903","source":{"kind":"arxiv","id":"2606.07435","version":1},"attestation_state":"computed","paper":{"title":"The Lipreading Gap: Do VSR Models Perceive Visual Speech Like Human Lipreaders?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Naomi Harte, Rishabh Jain","submitted_at":"2026-06-05T16:33:30Z","abstract_excerpt":"Visual speech recognition (VSR) models now surpass human lipreaders on benchmarks, but do such gains establish human-like visual speech perception? To explore this, we compare three VSR systems with human baselines on the MaFI word-level lipreading dataset using word, character, phoneme, and viseme-level metrics. Although models achieve higher overall accuracy, they succeed and fail on different words than humans. A text-only n-gram baseline given only a few initial phonemes rivals human lipreading. VSR word-level errors are consistently better explained by training word frequency than by the "},"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":"2606.07435","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-05T16:33:30Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"3bc15f57f34dcb5a4a750302ed54af905175cf4cbae42042b0ca2d2e37948c20","abstract_canon_sha256":"192914aeb0d65652ae0f4c21253d0ff96219d20efc3a83b711d7b1cc6aec4bba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:05:26.933095Z","signature_b64":"ST4RcyzotXPkJfo62ti1Q6zvaXk+Izx0BEx7/+fURW/o8PHNYGxlqOtaC4j5IQkimgUyon6lLS9/qqtv5vpmDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7df2aa6843d2a43d5f72d2c9460e1247079a808aa589989a214d173fea0e2903","last_reissued_at":"2026-06-08T01:05:26.932229Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:05:26.932229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Lipreading Gap: Do VSR Models Perceive Visual Speech Like Human Lipreaders?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Naomi Harte, Rishabh Jain","submitted_at":"2026-06-05T16:33:30Z","abstract_excerpt":"Visual speech recognition (VSR) models now surpass human lipreaders on benchmarks, but do such gains establish human-like visual speech perception? To explore this, we compare three VSR systems with human baselines on the MaFI word-level lipreading dataset using word, character, phoneme, and viseme-level metrics. Although models achieve higher overall accuracy, they succeed and fail on different words than humans. A text-only n-gram baseline given only a few initial phonemes rivals human lipreading. VSR word-level errors are consistently better explained by training word frequency than by the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.07435","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/2606.07435/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":"2606.07435","created_at":"2026-06-08T01:05:26.932383+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.07435v1","created_at":"2026-06-08T01:05:26.932383+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.07435","created_at":"2026-06-08T01:05:26.932383+00:00"},{"alias_kind":"pith_short_12","alias_value":"PXZKU2CD2KSD","created_at":"2026-06-08T01:05:26.932383+00:00"},{"alias_kind":"pith_short_16","alias_value":"PXZKU2CD2KSD2X3S","created_at":"2026-06-08T01:05:26.932383+00:00"},{"alias_kind":"pith_short_8","alias_value":"PXZKU2CD","created_at":"2026-06-08T01:05:26.932383+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.07435","citing_title":"The Lipreading Gap: Do VSR Models Perceive Visual Speech Like Human Lipreaders?","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4","json":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4.json","graph_json":"https://pith.science/api/pith-number/PXZKU2CD2KSD2X3S2LEUMDQSI4/graph.json","events_json":"https://pith.science/api/pith-number/PXZKU2CD2KSD2X3S2LEUMDQSI4/events.json","paper":"https://pith.science/paper/PXZKU2CD"},"agent_actions":{"view_html":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4","download_json":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4.json","view_paper":"https://pith.science/paper/PXZKU2CD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.07435&json=true","fetch_graph":"https://pith.science/api/pith-number/PXZKU2CD2KSD2X3S2LEUMDQSI4/graph.json","fetch_events":"https://pith.science/api/pith-number/PXZKU2CD2KSD2X3S2LEUMDQSI4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4/action/storage_attestation","attest_author":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4/action/author_attestation","sign_citation":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4/action/citation_signature","submit_replication":"https://pith.science/pith/PXZKU2CD2KSD2X3S2LEUMDQSI4/action/replication_record"}},"created_at":"2026-06-08T01:05:26.932383+00:00","updated_at":"2026-06-08T01:05:26.932383+00:00"}