{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:RDMSKUXLPUEDVGKFM2G7ZY27W4","short_pith_number":"pith:RDMSKUXL","schema_version":"1.0","canonical_sha256":"88d92552eb7d083a9945668dfce35fb73e306e403102bf96456cbdab09e641fe","source":{"kind":"arxiv","id":"2605.31530","version":1},"attestation_state":"computed","paper":{"title":"UNISON: A Unified Sound Generation and Editing Framework via Deep LLM Fusion","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Haoning Xu, Haoxuan Che, Huimeng Wang, Jiajun Deng, Jingran Su, Rui Liu, Tianzi Wang, Xunying Liu, Yaofang Liu, Zengrui Jin, Zhaoqing Li, Zhefan Rao","submitted_at":"2026-05-29T16:43:07Z","abstract_excerpt":"We present UNISON, a latent diffusion framework that unifies speech generation, sound generation, and audio editing within a single model. A single model handles text-to-audio, text-to-speech, zero-shot speaker cloning, mixed speech-and-sound generation, scene-level audio editing, speech-in-scene editing, and timed temporal composition, all of which share a single set of weights. Our architecture features two core designs: (1) Layer-wise deep LLM fusion, which injects hidden states from uniformly sampled layers of a frozen MLLM into corresponding MM-DiT blocks via learned projections, providin"},"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":"2605.31530","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2026-05-29T16:43:07Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"54e2633deecae30afac340d3c1f1cb29ddd38e00c81e099be63f50e1fae82bda","abstract_canon_sha256":"e70ec382c26713c2b395e87e0d317c15691b4cba90dc80384b73c600c14ab790"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T02:04:11.268272Z","signature_b64":"/hKwZi9T+vg9qQ6HNKpsyvkQ8K7nqLlRDfW7r+RR5trIVP4vf8CG4AKt3KstKIPO1TeGt6WvYDg7oD5mOqy/Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88d92552eb7d083a9945668dfce35fb73e306e403102bf96456cbdab09e641fe","last_reissued_at":"2026-06-01T02:04:11.267347Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T02:04:11.267347Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UNISON: A Unified Sound Generation and Editing Framework via Deep LLM Fusion","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"eess.AS","authors_text":"Haoning Xu, Haoxuan Che, Huimeng Wang, Jiajun Deng, Jingran Su, Rui Liu, Tianzi Wang, Xunying Liu, Yaofang Liu, Zengrui Jin, Zhaoqing Li, Zhefan Rao","submitted_at":"2026-05-29T16:43:07Z","abstract_excerpt":"We present UNISON, a latent diffusion framework that unifies speech generation, sound generation, and audio editing within a single model. A single model handles text-to-audio, text-to-speech, zero-shot speaker cloning, mixed speech-and-sound generation, scene-level audio editing, speech-in-scene editing, and timed temporal composition, all of which share a single set of weights. Our architecture features two core designs: (1) Layer-wise deep LLM fusion, which injects hidden states from uniformly sampled layers of a frozen MLLM into corresponding MM-DiT blocks via learned projections, providin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.31530","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/2605.31530/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":"2605.31530","created_at":"2026-06-01T02:04:11.267486+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.31530v1","created_at":"2026-06-01T02:04:11.267486+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.31530","created_at":"2026-06-01T02:04:11.267486+00:00"},{"alias_kind":"pith_short_12","alias_value":"RDMSKUXLPUED","created_at":"2026-06-01T02:04:11.267486+00:00"},{"alias_kind":"pith_short_16","alias_value":"RDMSKUXLPUEDVGKF","created_at":"2026-06-01T02:04:11.267486+00:00"},{"alias_kind":"pith_short_8","alias_value":"RDMSKUXL","created_at":"2026-06-01T02:04:11.267486+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/RDMSKUXLPUEDVGKFM2G7ZY27W4","json":"https://pith.science/pith/RDMSKUXLPUEDVGKFM2G7ZY27W4.json","graph_json":"https://pith.science/api/pith-number/RDMSKUXLPUEDVGKFM2G7ZY27W4/graph.json","events_json":"https://pith.science/api/pith-number/RDMSKUXLPUEDVGKFM2G7ZY27W4/events.json","paper":"https://pith.science/paper/RDMSKUXL"},"agent_actions":{"view_html":"https://pith.science/pith/RDMSKUXLPUEDVGKFM2G7ZY27W4","download_json":"https://pith.science/pith/RDMSKUXLPUEDVGKFM2G7ZY27W4.json","view_paper":"https://pith.science/paper/RDMSKUXL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.31530&json=true","fetch_graph":"https://pith.science/api/pith-number/RDMSKUXLPUEDVGKFM2G7ZY27W4/graph.json","fetch_events":"https://pith.science/api/pith-number/RDMSKUXLPUEDVGKFM2G7ZY27W4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RDMSKUXLPUEDVGKFM2G7ZY27W4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RDMSKUXLPUEDVGKFM2G7ZY27W4/action/storage_attestation","attest_author":"https://pith.science/pith/RDMSKUXLPUEDVGKFM2G7ZY27W4/action/author_attestation","sign_citation":"https://pith.science/pith/RDMSKUXLPUEDVGKFM2G7ZY27W4/action/citation_signature","submit_replication":"https://pith.science/pith/RDMSKUXLPUEDVGKFM2G7ZY27W4/action/replication_record"}},"created_at":"2026-06-01T02:04:11.267486+00:00","updated_at":"2026-06-01T02:04:11.267486+00:00"}