{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VM5YU2FGKAEEDOZZGYK5GQ4AA7","short_pith_number":"pith:VM5YU2FG","schema_version":"1.0","canonical_sha256":"ab3b8a68a6500841bb393615d3438007cb48a8eacde87e7fad48f9a243280b82","source":{"kind":"arxiv","id":"2506.18729","version":2},"attestation_state":"computed","paper":{"title":"MuseControlLite: Multifunctional Music Generation with Lightweight Conditioners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Bo-Rui Chen, Fang-Duo Tsai, Hao-Chung Cheng, Sheng-Ping Yang, Shih-Lun Wu, Weijaw Lee, Yi-Hsuan Yang","submitted_at":"2025-06-23T15:08:03Z","abstract_excerpt":"We propose MuseControlLite, a lightweight mechanism designed to fine-tune text-to-music generation models for precise conditioning using various time-varying musical attributes and reference audio signals. The key finding is that positional embeddings, which have been seldom used by text-to-music generation models in the conditioner for text conditions, are critical when the condition of interest is a function of time. Using melody control as an example, our experiments show that simply adding rotary positional embeddings to the decoupled cross-attention layers increases control accuracy from "},"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":"2506.18729","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-06-23T15:08:03Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"e9c0bc0260abc81da9d7021fcb1d31c2dfda8e78e138ede7ee7884bc2f5efb5c","abstract_canon_sha256":"cb5e7fc03b9bbbbabc9c67c97f9136ae901e6db833a62f68cbd01022e1335b86"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:33.520442Z","signature_b64":"LhCxjJ78CuDzEdSWI++3da88pJ5kny/Iilzzm19+U70i8okwtxrDqUR26rbcwmrt7dSliB/TO2PnsJOwoouxBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ab3b8a68a6500841bb393615d3438007cb48a8eacde87e7fad48f9a243280b82","last_reissued_at":"2026-07-05T11:26:33.519964Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:33.519964Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MuseControlLite: Multifunctional Music Generation with Lightweight Conditioners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Bo-Rui Chen, Fang-Duo Tsai, Hao-Chung Cheng, Sheng-Ping Yang, Shih-Lun Wu, Weijaw Lee, Yi-Hsuan Yang","submitted_at":"2025-06-23T15:08:03Z","abstract_excerpt":"We propose MuseControlLite, a lightweight mechanism designed to fine-tune text-to-music generation models for precise conditioning using various time-varying musical attributes and reference audio signals. The key finding is that positional embeddings, which have been seldom used by text-to-music generation models in the conditioner for text conditions, are critical when the condition of interest is a function of time. Using melody control as an example, our experiments show that simply adding rotary positional embeddings to the decoupled cross-attention layers increases control accuracy from "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18729","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/2506.18729/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":"2506.18729","created_at":"2026-07-05T11:26:33.520023+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.18729v2","created_at":"2026-07-05T11:26:33.520023+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18729","created_at":"2026-07-05T11:26:33.520023+00:00"},{"alias_kind":"pith_short_12","alias_value":"VM5YU2FGKAEE","created_at":"2026-07-05T11:26:33.520023+00:00"},{"alias_kind":"pith_short_16","alias_value":"VM5YU2FGKAEEDOZZ","created_at":"2026-07-05T11:26:33.520023+00:00"},{"alias_kind":"pith_short_8","alias_value":"VM5YU2FG","created_at":"2026-07-05T11:26:33.520023+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.31053","citing_title":"AnchorSteer: Self-Discovered Concept Injection for Structure-Preserving Music Editing","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22717","citing_title":"Live Music Diffusion Models: Efficient Fine-Tuning and Post-Training of Interactive Diffusion Music Generators","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7","json":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7.json","graph_json":"https://pith.science/api/pith-number/VM5YU2FGKAEEDOZZGYK5GQ4AA7/graph.json","events_json":"https://pith.science/api/pith-number/VM5YU2FGKAEEDOZZGYK5GQ4AA7/events.json","paper":"https://pith.science/paper/VM5YU2FG"},"agent_actions":{"view_html":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7","download_json":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7.json","view_paper":"https://pith.science/paper/VM5YU2FG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.18729&json=true","fetch_graph":"https://pith.science/api/pith-number/VM5YU2FGKAEEDOZZGYK5GQ4AA7/graph.json","fetch_events":"https://pith.science/api/pith-number/VM5YU2FGKAEEDOZZGYK5GQ4AA7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7/action/storage_attestation","attest_author":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7/action/author_attestation","sign_citation":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7/action/citation_signature","submit_replication":"https://pith.science/pith/VM5YU2FGKAEEDOZZGYK5GQ4AA7/action/replication_record"}},"created_at":"2026-07-05T11:26:33.520023+00:00","updated_at":"2026-07-05T11:26:33.520023+00:00"}