{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ETSPL4NI4U2OBFQHME35OPF66Y","short_pith_number":"pith:ETSPL4NI","schema_version":"1.0","canonical_sha256":"24e4f5f1a8e534e096076137d73cbef6364b9009f14c4686fb18f9cadb290ac2","source":{"kind":"arxiv","id":"2505.13270","version":1},"attestation_state":"computed","paper":{"title":"Distilling a speech and music encoder with task arithmetic","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Eng Siong Chng, Fabian Ritter-Gutierrez, Hung-yi Lee, Jeremy H.M Wong, Jui-Chiang Wei, Nancy F. Chen, Yi-Cheng Lin","submitted_at":"2025-05-19T15:51:53Z","abstract_excerpt":"Despite the progress in self-supervised learning (SSL) for speech and music, existing models treat these domains separately, limiting their capacity for unified audio understanding. A unified model is desirable for applications that require general representations, e.g. audio large language models. Nonetheless, directly training a general model for speech and music is computationally expensive. Knowledge Distillation of teacher ensembles may be a natural solution, but we posit that decoupling the distillation of the speech and music SSL models allows for more flexibility. Thus, we propose to l"},"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":"2505.13270","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-05-19T15:51:53Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"b0163be3205b91f063b3defb8b17b686cc8a34d2cb15fdb62ade80f4a077e65e","abstract_canon_sha256":"5370a4d5923a5c43e5f4d93f75ed9d490b4c9b28b792031a86814d851bc51f41"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:27.573186Z","signature_b64":"we0MF2MCvuFhHdlJcvlHAXxl8zfR1bTOYwr7dCN7yB5pl4o3t43yin3tJOaJ5FNW5QsrbCpZJB5Jv2mNmSATBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24e4f5f1a8e534e096076137d73cbef6364b9009f14c4686fb18f9cadb290ac2","last_reissued_at":"2026-07-05T11:05:27.572731Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:27.572731Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distilling a speech and music encoder with task arithmetic","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Eng Siong Chng, Fabian Ritter-Gutierrez, Hung-yi Lee, Jeremy H.M Wong, Jui-Chiang Wei, Nancy F. Chen, Yi-Cheng Lin","submitted_at":"2025-05-19T15:51:53Z","abstract_excerpt":"Despite the progress in self-supervised learning (SSL) for speech and music, existing models treat these domains separately, limiting their capacity for unified audio understanding. A unified model is desirable for applications that require general representations, e.g. audio large language models. Nonetheless, directly training a general model for speech and music is computationally expensive. Knowledge Distillation of teacher ensembles may be a natural solution, but we posit that decoupling the distillation of the speech and music SSL models allows for more flexibility. Thus, we propose to l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.13270","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/2505.13270/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":"2505.13270","created_at":"2026-07-05T11:05:27.572790+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.13270v1","created_at":"2026-07-05T11:05:27.572790+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.13270","created_at":"2026-07-05T11:05:27.572790+00:00"},{"alias_kind":"pith_short_12","alias_value":"ETSPL4NI4U2O","created_at":"2026-07-05T11:05:27.572790+00:00"},{"alias_kind":"pith_short_16","alias_value":"ETSPL4NI4U2OBFQH","created_at":"2026-07-05T11:05:27.572790+00:00"},{"alias_kind":"pith_short_8","alias_value":"ETSPL4NI","created_at":"2026-07-05T11:05:27.572790+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/ETSPL4NI4U2OBFQHME35OPF66Y","json":"https://pith.science/pith/ETSPL4NI4U2OBFQHME35OPF66Y.json","graph_json":"https://pith.science/api/pith-number/ETSPL4NI4U2OBFQHME35OPF66Y/graph.json","events_json":"https://pith.science/api/pith-number/ETSPL4NI4U2OBFQHME35OPF66Y/events.json","paper":"https://pith.science/paper/ETSPL4NI"},"agent_actions":{"view_html":"https://pith.science/pith/ETSPL4NI4U2OBFQHME35OPF66Y","download_json":"https://pith.science/pith/ETSPL4NI4U2OBFQHME35OPF66Y.json","view_paper":"https://pith.science/paper/ETSPL4NI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.13270&json=true","fetch_graph":"https://pith.science/api/pith-number/ETSPL4NI4U2OBFQHME35OPF66Y/graph.json","fetch_events":"https://pith.science/api/pith-number/ETSPL4NI4U2OBFQHME35OPF66Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ETSPL4NI4U2OBFQHME35OPF66Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ETSPL4NI4U2OBFQHME35OPF66Y/action/storage_attestation","attest_author":"https://pith.science/pith/ETSPL4NI4U2OBFQHME35OPF66Y/action/author_attestation","sign_citation":"https://pith.science/pith/ETSPL4NI4U2OBFQHME35OPF66Y/action/citation_signature","submit_replication":"https://pith.science/pith/ETSPL4NI4U2OBFQHME35OPF66Y/action/replication_record"}},"created_at":"2026-07-05T11:05:27.572790+00:00","updated_at":"2026-07-05T11:05:27.572790+00:00"}