{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6NOWCFC2IC6CEXKBSIQ4256ALM","short_pith_number":"pith:6NOWCFC2","schema_version":"1.0","canonical_sha256":"f35d61145a40bc225d419221cd77c05b33457df3ebbeb61654fcb031beb38c3b","source":{"kind":"arxiv","id":"2509.06598","version":1},"attestation_state":"computed","paper":{"title":"Integrating Spatial and Semantic Embeddings for Stereo Sound Event Localization in Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.IV","eess.SP"],"primary_cat":"eess.AS","authors_text":"Davide Berghi, Philip J. B. Jackson","submitted_at":"2025-09-08T12:07:32Z","abstract_excerpt":"In this study, we address the multimodal task of stereo sound event localization and detection with source distance estimation (3D SELD) in regular video content. 3D SELD is a complex task that combines temporal event classification with spatial localization, requiring reasoning across spatial, temporal, and semantic dimensions. The last is arguably the most challenging to model. Traditional SELD approaches typically rely on multichannel input, limiting their capacity to benefit from large-scale pre-training due to data constraints. To overcome this, we enhance a standard SELD architecture 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":"2509.06598","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2025-09-08T12:07:32Z","cross_cats_sorted":["cs.AI","cs.LG","eess.IV","eess.SP"],"title_canon_sha256":"45a38a90c7d553c0f8d71afe528072da176b7ef3afc1173fcedcfed97e95157d","abstract_canon_sha256":"a141a91e59a5f8f925430dd6a9c02e3e2df2591e82dfa329c08227ed543f99b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:39.012564Z","signature_b64":"qCk5rMCcVx2mqHLn0/vvoH7kYyBe49knxMeB3MJ4lgcyiCypEcjXTOiJJXVNOHdRIIQmYCG7Cz+r8QyI6mJrAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f35d61145a40bc225d419221cd77c05b33457df3ebbeb61654fcb031beb38c3b","last_reissued_at":"2026-07-05T12:06:39.012099Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:39.012099Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Integrating Spatial and Semantic Embeddings for Stereo Sound Event Localization in Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.IV","eess.SP"],"primary_cat":"eess.AS","authors_text":"Davide Berghi, Philip J. B. Jackson","submitted_at":"2025-09-08T12:07:32Z","abstract_excerpt":"In this study, we address the multimodal task of stereo sound event localization and detection with source distance estimation (3D SELD) in regular video content. 3D SELD is a complex task that combines temporal event classification with spatial localization, requiring reasoning across spatial, temporal, and semantic dimensions. The last is arguably the most challenging to model. Traditional SELD approaches typically rely on multichannel input, limiting their capacity to benefit from large-scale pre-training due to data constraints. To overcome this, we enhance a standard SELD architecture wit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.06598","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/2509.06598/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":"2509.06598","created_at":"2026-07-05T12:06:39.012154+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.06598v1","created_at":"2026-07-05T12:06:39.012154+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.06598","created_at":"2026-07-05T12:06:39.012154+00:00"},{"alias_kind":"pith_short_12","alias_value":"6NOWCFC2IC6C","created_at":"2026-07-05T12:06:39.012154+00:00"},{"alias_kind":"pith_short_16","alias_value":"6NOWCFC2IC6CEXKB","created_at":"2026-07-05T12:06:39.012154+00:00"},{"alias_kind":"pith_short_8","alias_value":"6NOWCFC2","created_at":"2026-07-05T12:06:39.012154+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/6NOWCFC2IC6CEXKBSIQ4256ALM","json":"https://pith.science/pith/6NOWCFC2IC6CEXKBSIQ4256ALM.json","graph_json":"https://pith.science/api/pith-number/6NOWCFC2IC6CEXKBSIQ4256ALM/graph.json","events_json":"https://pith.science/api/pith-number/6NOWCFC2IC6CEXKBSIQ4256ALM/events.json","paper":"https://pith.science/paper/6NOWCFC2"},"agent_actions":{"view_html":"https://pith.science/pith/6NOWCFC2IC6CEXKBSIQ4256ALM","download_json":"https://pith.science/pith/6NOWCFC2IC6CEXKBSIQ4256ALM.json","view_paper":"https://pith.science/paper/6NOWCFC2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.06598&json=true","fetch_graph":"https://pith.science/api/pith-number/6NOWCFC2IC6CEXKBSIQ4256ALM/graph.json","fetch_events":"https://pith.science/api/pith-number/6NOWCFC2IC6CEXKBSIQ4256ALM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6NOWCFC2IC6CEXKBSIQ4256ALM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6NOWCFC2IC6CEXKBSIQ4256ALM/action/storage_attestation","attest_author":"https://pith.science/pith/6NOWCFC2IC6CEXKBSIQ4256ALM/action/author_attestation","sign_citation":"https://pith.science/pith/6NOWCFC2IC6CEXKBSIQ4256ALM/action/citation_signature","submit_replication":"https://pith.science/pith/6NOWCFC2IC6CEXKBSIQ4256ALM/action/replication_record"}},"created_at":"2026-07-05T12:06:39.012154+00:00","updated_at":"2026-07-05T12:06:39.012154+00:00"}