{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HJJK3ZKOOLN6IHI5CLV57MWRLO","short_pith_number":"pith:HJJK3ZKO","schema_version":"1.0","canonical_sha256":"3a52ade54e72dbe41d1d12ebdfb2d15b923cb9d384881a25bda9184244d5cc0d","source":{"kind":"arxiv","id":"2507.17297","version":1},"attestation_state":"computed","paper":{"title":"On Temporal Guidance and Iterative Refinement in Audio Source Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Florian Schmid, Gerhard Widmer, Jonathan Greif, Paul Primus, Tobias Morocutti","submitted_at":"2025-07-23T07:58:28Z","abstract_excerpt":"Spatial semantic segmentation of sound scenes (S5) involves the accurate identification of active sound classes and the precise separation of their sources from complex acoustic mixtures. Conventional systems rely on a two-stage pipeline - audio tagging followed by label-conditioned source separation - but are often constrained by the absence of fine-grained temporal information critical for effective separation. In this work, we address this limitation by introducing a novel approach for S5 that enhances the synergy between the event detection and source separation stages. Our key contributio"},"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":"2507.17297","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2025-07-23T07:58:28Z","cross_cats_sorted":["cs.AI","cs.LG","eess.AS"],"title_canon_sha256":"d310b0e540c6429c2993fa64c76778ddc7b2c8181813957491ae4b33216498dc","abstract_canon_sha256":"8a388a7a932bfdafd11efd6fa3335cc265236472d4a972f2650c4ea7386188c3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:05.427058Z","signature_b64":"W9Ya+AUrnBM0cTFEWmkDyayo9wPD9h9+CFWmp9gZ04e3dy+iO7b2f6mvoSx/JplnZeZJxnvu5HxdYC6zeFzHBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a52ade54e72dbe41d1d12ebdfb2d15b923cb9d384881a25bda9184244d5cc0d","last_reissued_at":"2026-07-05T11:42:05.426559Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:05.426559Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Temporal Guidance and Iterative Refinement in Audio Source Separation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Florian Schmid, Gerhard Widmer, Jonathan Greif, Paul Primus, Tobias Morocutti","submitted_at":"2025-07-23T07:58:28Z","abstract_excerpt":"Spatial semantic segmentation of sound scenes (S5) involves the accurate identification of active sound classes and the precise separation of their sources from complex acoustic mixtures. Conventional systems rely on a two-stage pipeline - audio tagging followed by label-conditioned source separation - but are often constrained by the absence of fine-grained temporal information critical for effective separation. In this work, we address this limitation by introducing a novel approach for S5 that enhances the synergy between the event detection and source separation stages. Our key contributio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.17297","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/2507.17297/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":"2507.17297","created_at":"2026-07-05T11:42:05.426622+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.17297v1","created_at":"2026-07-05T11:42:05.426622+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.17297","created_at":"2026-07-05T11:42:05.426622+00:00"},{"alias_kind":"pith_short_12","alias_value":"HJJK3ZKOOLN6","created_at":"2026-07-05T11:42:05.426622+00:00"},{"alias_kind":"pith_short_16","alias_value":"HJJK3ZKOOLN6IHI5","created_at":"2026-07-05T11:42:05.426622+00:00"},{"alias_kind":"pith_short_8","alias_value":"HJJK3ZKO","created_at":"2026-07-05T11:42:05.426622+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/HJJK3ZKOOLN6IHI5CLV57MWRLO","json":"https://pith.science/pith/HJJK3ZKOOLN6IHI5CLV57MWRLO.json","graph_json":"https://pith.science/api/pith-number/HJJK3ZKOOLN6IHI5CLV57MWRLO/graph.json","events_json":"https://pith.science/api/pith-number/HJJK3ZKOOLN6IHI5CLV57MWRLO/events.json","paper":"https://pith.science/paper/HJJK3ZKO"},"agent_actions":{"view_html":"https://pith.science/pith/HJJK3ZKOOLN6IHI5CLV57MWRLO","download_json":"https://pith.science/pith/HJJK3ZKOOLN6IHI5CLV57MWRLO.json","view_paper":"https://pith.science/paper/HJJK3ZKO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.17297&json=true","fetch_graph":"https://pith.science/api/pith-number/HJJK3ZKOOLN6IHI5CLV57MWRLO/graph.json","fetch_events":"https://pith.science/api/pith-number/HJJK3ZKOOLN6IHI5CLV57MWRLO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HJJK3ZKOOLN6IHI5CLV57MWRLO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HJJK3ZKOOLN6IHI5CLV57MWRLO/action/storage_attestation","attest_author":"https://pith.science/pith/HJJK3ZKOOLN6IHI5CLV57MWRLO/action/author_attestation","sign_citation":"https://pith.science/pith/HJJK3ZKOOLN6IHI5CLV57MWRLO/action/citation_signature","submit_replication":"https://pith.science/pith/HJJK3ZKOOLN6IHI5CLV57MWRLO/action/replication_record"}},"created_at":"2026-07-05T11:42:05.426622+00:00","updated_at":"2026-07-05T11:42:05.426622+00:00"}