{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7TK4BKAJ53BPG7GYQLE6TPIXPN","short_pith_number":"pith:7TK4BKAJ","schema_version":"1.0","canonical_sha256":"fcd5c0a809eec2f37cd882c9e9bd177b61ef2aef9ecd763ae9662e6f691a9459","source":{"kind":"arxiv","id":"2606.29632","version":1},"attestation_state":"computed","paper":{"title":"VIB-AVSR: Variational Information Bottleneck for Noise-Robust LLM-Based 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, Navlika Singh, Piyush Arora, Stavros Petridis, Umberto Cappellazzo","submitted_at":"2026-06-28T22:31:28Z","abstract_excerpt":"Audio-Visual Speech Recognition takes two input modalities, acoustic and visual streams, where visual information from lip movements aids recognition when audio is noisy. Recently, LLM-based AVSR models have emerged as a promising paradigm by connecting pre-trained audio-visual encoders to an LLM, achieving strong results in clean conditions. However, these models are predominantly optimized for clean acoustic conditions, with limited attention to making the LLM backbone robust to noise. No explicit mechanism is employed to produce stable representations under corrupted audio, leading to perfo"},"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":"2606.29632","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2026-06-28T22:31:28Z","cross_cats_sorted":["cs.CV","cs.SD"],"title_canon_sha256":"57ac464070f3faef022bb3fa7cf015a6fdc891d90f12c01b00c978909945a7fe","abstract_canon_sha256":"d67f34d71a80a563562b3b89cbb86736c8de0fcefac6f961969b8e97555e7b57"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:18:20.359340Z","signature_b64":"FCL7YDSkg0S0xPlGAbtHYcoW+juKq5J0GEj7SO1iAVOdLnqDs91SAWicWi/BAL/O+e5Uw68OWzw2Pb4wWJPnDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcd5c0a809eec2f37cd882c9e9bd177b61ef2aef9ecd763ae9662e6f691a9459","last_reissued_at":"2026-06-30T01:18:20.358810Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:18:20.358810Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VIB-AVSR: Variational Information Bottleneck for Noise-Robust LLM-Based 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, Navlika Singh, Piyush Arora, Stavros Petridis, Umberto Cappellazzo","submitted_at":"2026-06-28T22:31:28Z","abstract_excerpt":"Audio-Visual Speech Recognition takes two input modalities, acoustic and visual streams, where visual information from lip movements aids recognition when audio is noisy. Recently, LLM-based AVSR models have emerged as a promising paradigm by connecting pre-trained audio-visual encoders to an LLM, achieving strong results in clean conditions. However, these models are predominantly optimized for clean acoustic conditions, with limited attention to making the LLM backbone robust to noise. No explicit mechanism is employed to produce stable representations under corrupted audio, leading to perfo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29632","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/2606.29632/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":"2606.29632","created_at":"2026-06-30T01:18:20.358892+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.29632v1","created_at":"2026-06-30T01:18:20.358892+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29632","created_at":"2026-06-30T01:18:20.358892+00:00"},{"alias_kind":"pith_short_12","alias_value":"7TK4BKAJ53BP","created_at":"2026-06-30T01:18:20.358892+00:00"},{"alias_kind":"pith_short_16","alias_value":"7TK4BKAJ53BPG7GY","created_at":"2026-06-30T01:18:20.358892+00:00"},{"alias_kind":"pith_short_8","alias_value":"7TK4BKAJ","created_at":"2026-06-30T01:18:20.358892+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/7TK4BKAJ53BPG7GYQLE6TPIXPN","json":"https://pith.science/pith/7TK4BKAJ53BPG7GYQLE6TPIXPN.json","graph_json":"https://pith.science/api/pith-number/7TK4BKAJ53BPG7GYQLE6TPIXPN/graph.json","events_json":"https://pith.science/api/pith-number/7TK4BKAJ53BPG7GYQLE6TPIXPN/events.json","paper":"https://pith.science/paper/7TK4BKAJ"},"agent_actions":{"view_html":"https://pith.science/pith/7TK4BKAJ53BPG7GYQLE6TPIXPN","download_json":"https://pith.science/pith/7TK4BKAJ53BPG7GYQLE6TPIXPN.json","view_paper":"https://pith.science/paper/7TK4BKAJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.29632&json=true","fetch_graph":"https://pith.science/api/pith-number/7TK4BKAJ53BPG7GYQLE6TPIXPN/graph.json","fetch_events":"https://pith.science/api/pith-number/7TK4BKAJ53BPG7GYQLE6TPIXPN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7TK4BKAJ53BPG7GYQLE6TPIXPN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7TK4BKAJ53BPG7GYQLE6TPIXPN/action/storage_attestation","attest_author":"https://pith.science/pith/7TK4BKAJ53BPG7GYQLE6TPIXPN/action/author_attestation","sign_citation":"https://pith.science/pith/7TK4BKAJ53BPG7GYQLE6TPIXPN/action/citation_signature","submit_replication":"https://pith.science/pith/7TK4BKAJ53BPG7GYQLE6TPIXPN/action/replication_record"}},"created_at":"2026-06-30T01:18:20.358892+00:00","updated_at":"2026-06-30T01:18:20.358892+00:00"}