{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BOTOQJTXVYBZIJSVALVK5BGUMR","short_pith_number":"pith:BOTOQJTX","schema_version":"1.0","canonical_sha256":"0ba6e82677ae0394265502eaae84d46443998ed6b11db528260b71386d3e40fd","source":{"kind":"arxiv","id":"2608.08757","version":1},"attestation_state":"computed","paper":{"title":"Steering dense music retrieval with open-vocabulary concept discovery","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Alain Riou, Elio Quinton, Gy\\\"orgy Fazekas, Julien Guinot","submitted_at":"2026-08-09T15:16:58Z","abstract_excerpt":"Controllable music retrieval lets users find music that is, for example, more ambient, less distorted, or without guitar while preserving the other semantic content of an original seed query. Sparse autoencoders (SAEs) are a promising interface for this kind of concept-level control, but a key problem remains: given a free-form text concept, which sparse features should be edited? In shared multimodal embedding spaces, standard attribution methods often select neurons that match the concept's wording but not the audio examples that express it. This leads to weak or unstable edits: relevant fea"},"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":"2608.08757","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-08-09T15:16:58Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"641c7af4bf654300c1640d2b3b6755dc07c54d2edd8b90699e588a0492a134e6","abstract_canon_sha256":"3da0a667a823b78aa40ddd4b5dc8d1207c7b81ad0b667d850430c30553cad51d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T01:25:15.051140Z","signature_b64":"AV7/pT7CA3ZtU5HPE1myjgJKml52/DOub1NyDafwID1v9EcoEXMtV6R/6rtWsDZUP40HH29zqSSKMZMZPJOKAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0ba6e82677ae0394265502eaae84d46443998ed6b11db528260b71386d3e40fd","last_reissued_at":"2026-08-11T01:25:15.048397Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T01:25:15.048397Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Steering dense music retrieval with open-vocabulary concept discovery","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Alain Riou, Elio Quinton, Gy\\\"orgy Fazekas, Julien Guinot","submitted_at":"2026-08-09T15:16:58Z","abstract_excerpt":"Controllable music retrieval lets users find music that is, for example, more ambient, less distorted, or without guitar while preserving the other semantic content of an original seed query. Sparse autoencoders (SAEs) are a promising interface for this kind of concept-level control, but a key problem remains: given a free-form text concept, which sparse features should be edited? In shared multimodal embedding spaces, standard attribution methods often select neurons that match the concept's wording but not the audio examples that express it. This leads to weak or unstable edits: relevant fea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.08757","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/2608.08757/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":"2608.08757","created_at":"2026-08-11T01:25:15.049479+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.08757v1","created_at":"2026-08-11T01:25:15.049479+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.08757","created_at":"2026-08-11T01:25:15.049479+00:00"},{"alias_kind":"pith_short_12","alias_value":"BOTOQJTXVYBZ","created_at":"2026-08-11T01:25:15.049479+00:00"},{"alias_kind":"pith_short_16","alias_value":"BOTOQJTXVYBZIJSV","created_at":"2026-08-11T01:25:15.049479+00:00"},{"alias_kind":"pith_short_8","alias_value":"BOTOQJTX","created_at":"2026-08-11T01:25:15.049479+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.08757","citing_title":"Steering dense music retrieval with open-vocabulary concept discovery","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR","json":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR.json","graph_json":"https://pith.science/api/pith-number/BOTOQJTXVYBZIJSVALVK5BGUMR/graph.json","events_json":"https://pith.science/api/pith-number/BOTOQJTXVYBZIJSVALVK5BGUMR/events.json","paper":"https://pith.science/paper/BOTOQJTX"},"agent_actions":{"view_html":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR","download_json":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR.json","view_paper":"https://pith.science/paper/BOTOQJTX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.08757&json=true","fetch_graph":"https://pith.science/api/pith-number/BOTOQJTXVYBZIJSVALVK5BGUMR/graph.json","fetch_events":"https://pith.science/api/pith-number/BOTOQJTXVYBZIJSVALVK5BGUMR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR/action/storage_attestation","attest_author":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR/action/author_attestation","sign_citation":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR/action/citation_signature","submit_replication":"https://pith.science/pith/BOTOQJTXVYBZIJSVALVK5BGUMR/action/replication_record"}},"created_at":"2026-08-11T01:25:15.049479+00:00","updated_at":"2026-08-11T01:25:15.049479+00:00"}