{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PEKCP2SBWBR6ERFH3XAZG5DNZ2","short_pith_number":"pith:PEKCP2SB","schema_version":"1.0","canonical_sha256":"791427ea41b063e244a7ddc193746dcea0a4b520dca4ea7b53206172be789170","source":{"kind":"arxiv","id":"2507.11967","version":1},"attestation_state":"computed","paper":{"title":"Language-Guided Contrastive Audio-Visual Masked Autoencoder with Automatically Generated Audio-Visual-Text Triplets from Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS","eess.IV"],"primary_cat":"cs.CV","authors_text":"Hokuto Munakata, Kazuhiro Saito, Shota Nakada, Tatsuya Komatsu, Yoshimitsu Aoki, Yuchi Ishikawa","submitted_at":"2025-07-16T06:58:14Z","abstract_excerpt":"In this paper, we propose Language-Guided Contrastive Audio-Visual Masked Autoencoders (LG-CAV-MAE) to improve audio-visual representation learning. LG-CAV-MAE integrates a pretrained text encoder into contrastive audio-visual masked autoencoders, enabling the model to learn across audio, visual and text modalities. To train LG-CAV-MAE, we introduce an automatic method to generate audio-visual-text triplets from unlabeled videos. We first generate frame-level captions using an image captioning model and then apply CLAP-based filtering to ensure strong alignment between audio and captions. This"},"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":"2507.11967","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-16T06:58:14Z","cross_cats_sorted":["eess.AS","eess.IV"],"title_canon_sha256":"2d15514503d4fd4b4a97b4a3481c51a9ea69c9ccce103de10bfdebe13e37341f","abstract_canon_sha256":"8ff0edbe3fb471d278ece21dec50ebe9e135efef37f37f25ba5f3f549d26d581"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:11.977438Z","signature_b64":"tUdAWs2dGhrSyPrek0d7URVdukdzRfiAtqH2OMrykSADKy0lwKmBoTJMgH1sWZgw5wZQgKQYJiwa3olhQOdAAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"791427ea41b063e244a7ddc193746dcea0a4b520dca4ea7b53206172be789170","last_reissued_at":"2026-07-05T11:38:11.976977Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:11.976977Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Language-Guided Contrastive Audio-Visual Masked Autoencoder with Automatically Generated Audio-Visual-Text Triplets from Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS","eess.IV"],"primary_cat":"cs.CV","authors_text":"Hokuto Munakata, Kazuhiro Saito, Shota Nakada, Tatsuya Komatsu, Yoshimitsu Aoki, Yuchi Ishikawa","submitted_at":"2025-07-16T06:58:14Z","abstract_excerpt":"In this paper, we propose Language-Guided Contrastive Audio-Visual Masked Autoencoders (LG-CAV-MAE) to improve audio-visual representation learning. LG-CAV-MAE integrates a pretrained text encoder into contrastive audio-visual masked autoencoders, enabling the model to learn across audio, visual and text modalities. To train LG-CAV-MAE, we introduce an automatic method to generate audio-visual-text triplets from unlabeled videos. We first generate frame-level captions using an image captioning model and then apply CLAP-based filtering to ensure strong alignment between audio and captions. This"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.11967","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/2507.11967/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":"2507.11967","created_at":"2026-07-05T11:38:11.977035+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.11967v1","created_at":"2026-07-05T11:38:11.977035+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.11967","created_at":"2026-07-05T11:38:11.977035+00:00"},{"alias_kind":"pith_short_12","alias_value":"PEKCP2SBWBR6","created_at":"2026-07-05T11:38:11.977035+00:00"},{"alias_kind":"pith_short_16","alias_value":"PEKCP2SBWBR6ERFH","created_at":"2026-07-05T11:38:11.977035+00:00"},{"alias_kind":"pith_short_8","alias_value":"PEKCP2SB","created_at":"2026-07-05T11:38:11.977035+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.11967","citing_title":"Language-Guided Contrastive Audio-Visual Masked Autoencoder with Automatically Generated Audio-Visual-Text Triplets from Videos","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2","json":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2.json","graph_json":"https://pith.science/api/pith-number/PEKCP2SBWBR6ERFH3XAZG5DNZ2/graph.json","events_json":"https://pith.science/api/pith-number/PEKCP2SBWBR6ERFH3XAZG5DNZ2/events.json","paper":"https://pith.science/paper/PEKCP2SB"},"agent_actions":{"view_html":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2","download_json":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2.json","view_paper":"https://pith.science/paper/PEKCP2SB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.11967&json=true","fetch_graph":"https://pith.science/api/pith-number/PEKCP2SBWBR6ERFH3XAZG5DNZ2/graph.json","fetch_events":"https://pith.science/api/pith-number/PEKCP2SBWBR6ERFH3XAZG5DNZ2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2/action/storage_attestation","attest_author":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2/action/author_attestation","sign_citation":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2/action/citation_signature","submit_replication":"https://pith.science/pith/PEKCP2SBWBR6ERFH3XAZG5DNZ2/action/replication_record"}},"created_at":"2026-07-05T11:38:11.977035+00:00","updated_at":"2026-07-05T11:38:11.977035+00:00"}