{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TSFQLRLVGYK6BHY6Y3DPPMEIC5","short_pith_number":"pith:TSFQLRLV","schema_version":"1.0","canonical_sha256":"9c8b05c5753615e09f1ec6c6f7b0881766729d5b2c5d7488fd14f1fb350bee34","source":{"kind":"arxiv","id":"2507.04959","version":2},"attestation_state":"computed","paper":{"title":"Hear-Your-Click: Interactive Object-Specific Video-to-Audio Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MM","cs.SD","eess.AS"],"primary_cat":"cs.CV","authors_text":"Jiasheng Lu, Keyu Fan, Peiwu Qin, Qingyang Shi, Xinyu Zhang, Yingshan Liang, Yiran Wang, Zhicheng Du","submitted_at":"2025-07-07T13:01:50Z","abstract_excerpt":"Video-to-audio (V2A) generation shows great potential in fields such as film production. Despite significant advances, current V2A methods relying on global video information struggle with complex scenes and generating audio tailored to specific objects. To address these limitations, we introduce Hear-Your-Click, an interactive V2A framework enabling users to generate sounds for specific objects by clicking on the frame. To achieve this, we propose Object-aware Contrastive Audio-Visual Fine-tuning (OCAV) with a Mask-guided Visual Encoder (MVE) to obtain object-level visual features aligned wit"},"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":"2507.04959","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-07T13:01:50Z","cross_cats_sorted":["cs.AI","cs.MM","cs.SD","eess.AS"],"title_canon_sha256":"a2e43f73610edfdda473aa82fc86634f3395010c1cf7ee93098b5e72a7fe72b6","abstract_canon_sha256":"15b0d212579bfe84883c50c2c6c8e3d3fcca84220abfa18e7ea22839a0f59a0c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:35:57.919407Z","signature_b64":"YLJ5zZ6816XHfFnNGaFdNKBxvyR+SHNfvHvq5hNw1zRn2XRZl7e0bDpXPdFq2NCxVDo51CIsEYpA9Ku/y2JsAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c8b05c5753615e09f1ec6c6f7b0881766729d5b2c5d7488fd14f1fb350bee34","last_reissued_at":"2026-07-05T11:35:57.918929Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:35:57.918929Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hear-Your-Click: Interactive Object-Specific Video-to-Audio Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MM","cs.SD","eess.AS"],"primary_cat":"cs.CV","authors_text":"Jiasheng Lu, Keyu Fan, Peiwu Qin, Qingyang Shi, Xinyu Zhang, Yingshan Liang, Yiran Wang, Zhicheng Du","submitted_at":"2025-07-07T13:01:50Z","abstract_excerpt":"Video-to-audio (V2A) generation shows great potential in fields such as film production. Despite significant advances, current V2A methods relying on global video information struggle with complex scenes and generating audio tailored to specific objects. To address these limitations, we introduce Hear-Your-Click, an interactive V2A framework enabling users to generate sounds for specific objects by clicking on the frame. To achieve this, we propose Object-aware Contrastive Audio-Visual Fine-tuning (OCAV) with a Mask-guided Visual Encoder (MVE) to obtain object-level visual features aligned wit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.04959","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/2507.04959/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":"2507.04959","created_at":"2026-07-05T11:35:57.918988+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.04959v2","created_at":"2026-07-05T11:35:57.918988+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.04959","created_at":"2026-07-05T11:35:57.918988+00:00"},{"alias_kind":"pith_short_12","alias_value":"TSFQLRLVGYK6","created_at":"2026-07-05T11:35:57.918988+00:00"},{"alias_kind":"pith_short_16","alias_value":"TSFQLRLVGYK6BHY6","created_at":"2026-07-05T11:35:57.918988+00:00"},{"alias_kind":"pith_short_8","alias_value":"TSFQLRLV","created_at":"2026-07-05T11:35:57.918988+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/TSFQLRLVGYK6BHY6Y3DPPMEIC5","json":"https://pith.science/pith/TSFQLRLVGYK6BHY6Y3DPPMEIC5.json","graph_json":"https://pith.science/api/pith-number/TSFQLRLVGYK6BHY6Y3DPPMEIC5/graph.json","events_json":"https://pith.science/api/pith-number/TSFQLRLVGYK6BHY6Y3DPPMEIC5/events.json","paper":"https://pith.science/paper/TSFQLRLV"},"agent_actions":{"view_html":"https://pith.science/pith/TSFQLRLVGYK6BHY6Y3DPPMEIC5","download_json":"https://pith.science/pith/TSFQLRLVGYK6BHY6Y3DPPMEIC5.json","view_paper":"https://pith.science/paper/TSFQLRLV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.04959&json=true","fetch_graph":"https://pith.science/api/pith-number/TSFQLRLVGYK6BHY6Y3DPPMEIC5/graph.json","fetch_events":"https://pith.science/api/pith-number/TSFQLRLVGYK6BHY6Y3DPPMEIC5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TSFQLRLVGYK6BHY6Y3DPPMEIC5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TSFQLRLVGYK6BHY6Y3DPPMEIC5/action/storage_attestation","attest_author":"https://pith.science/pith/TSFQLRLVGYK6BHY6Y3DPPMEIC5/action/author_attestation","sign_citation":"https://pith.science/pith/TSFQLRLVGYK6BHY6Y3DPPMEIC5/action/citation_signature","submit_replication":"https://pith.science/pith/TSFQLRLVGYK6BHY6Y3DPPMEIC5/action/replication_record"}},"created_at":"2026-07-05T11:35:57.918988+00:00","updated_at":"2026-07-05T11:35:57.918988+00:00"}