{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WU5H7LJYNWJLIJXCIV2NKBQYYJ","short_pith_number":"pith:WU5H7LJY","schema_version":"1.0","canonical_sha256":"b53a7fad386d92b426e24574d50618c2764bba9c23d64cce26c9e013f3522a2b","source":{"kind":"arxiv","id":"2312.06720","version":2},"attestation_state":"computed","paper":{"title":"Audio-Visual LLM for Video Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cihang Xie, Fangxun Shu, Hao Jiang, Lei Zhang","submitted_at":"2023-12-11T02:50:46Z","abstract_excerpt":"This paper presents Audio-Visual LLM, a Multimodal Large Language Model that takes both visual and auditory inputs for holistic video understanding. A key design is the modality-augmented training, which involves the integration of modality-specific tokens engineered to activate the appropriate visual and/or auditory encoder selectively. This mechanism is pivotal in enabling end-to-end joint training with video data at different modalities, including visual-only, audio-only, and audio-visual formats. Moreover, we introduce a high-quality video instruction dataset, derived from GPT-4. This data"},"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":"2312.06720","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-12-11T02:50:46Z","cross_cats_sorted":[],"title_canon_sha256":"9526c71988bb21d08cf67ddb73eeeba9f6ab8433d61b3e6adaeb0108b817b512","abstract_canon_sha256":"928620f092568c9cca18a74352dd8e4f08a3ea98bce9851334db67230db2a850"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:24:02.906607Z","signature_b64":"zEU+Yaszzd6EFglvz9PraLJgA3PZheOjUAma/c+xwju+WpTB4ycoA8cJxIO8cYHe3/xeKEIza8BpqawcGZLzAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b53a7fad386d92b426e24574d50618c2764bba9c23d64cce26c9e013f3522a2b","last_reissued_at":"2026-07-05T07:24:02.906153Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:24:02.906153Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Audio-Visual LLM for Video Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cihang Xie, Fangxun Shu, Hao Jiang, Lei Zhang","submitted_at":"2023-12-11T02:50:46Z","abstract_excerpt":"This paper presents Audio-Visual LLM, a Multimodal Large Language Model that takes both visual and auditory inputs for holistic video understanding. A key design is the modality-augmented training, which involves the integration of modality-specific tokens engineered to activate the appropriate visual and/or auditory encoder selectively. This mechanism is pivotal in enabling end-to-end joint training with video data at different modalities, including visual-only, audio-only, and audio-visual formats. Moreover, we introduce a high-quality video instruction dataset, derived from GPT-4. This data"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.06720","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/2312.06720/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":"2312.06720","created_at":"2026-07-05T07:24:02.906205+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.06720v2","created_at":"2026-07-05T07:24:02.906205+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.06720","created_at":"2026-07-05T07:24:02.906205+00:00"},{"alias_kind":"pith_short_12","alias_value":"WU5H7LJYNWJL","created_at":"2026-07-05T07:24:02.906205+00:00"},{"alias_kind":"pith_short_16","alias_value":"WU5H7LJYNWJLIJXC","created_at":"2026-07-05T07:24:02.906205+00:00"},{"alias_kind":"pith_short_8","alias_value":"WU5H7LJY","created_at":"2026-07-05T07:24:02.906205+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10533","citing_title":"Audio-Visual Exchange-Aware Token Pruning for Efficient Audio-Visual Captioning","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2510.18346","citing_title":"AV-Master: Dual-Path Comprehensive Perception Makes Better Audio-Visual Question Answering","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2501.01957","citing_title":"VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2503.21776","citing_title":"Video-R1: Reinforcing Video Reasoning in MLLMs","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2406.07476","citing_title":"VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2501.13106","citing_title":"VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding","ref_index":157,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ","json":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ.json","graph_json":"https://pith.science/api/pith-number/WU5H7LJYNWJLIJXCIV2NKBQYYJ/graph.json","events_json":"https://pith.science/api/pith-number/WU5H7LJYNWJLIJXCIV2NKBQYYJ/events.json","paper":"https://pith.science/paper/WU5H7LJY"},"agent_actions":{"view_html":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ","download_json":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ.json","view_paper":"https://pith.science/paper/WU5H7LJY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.06720&json=true","fetch_graph":"https://pith.science/api/pith-number/WU5H7LJYNWJLIJXCIV2NKBQYYJ/graph.json","fetch_events":"https://pith.science/api/pith-number/WU5H7LJYNWJLIJXCIV2NKBQYYJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ/action/storage_attestation","attest_author":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ/action/author_attestation","sign_citation":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ/action/citation_signature","submit_replication":"https://pith.science/pith/WU5H7LJYNWJLIJXCIV2NKBQYYJ/action/replication_record"}},"created_at":"2026-07-05T07:24:02.906205+00:00","updated_at":"2026-07-05T07:24:02.906205+00:00"}