{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KIXGXHKRPWOYOFZZXVF5WR4BLF","short_pith_number":"pith:KIXGXHKR","schema_version":"1.0","canonical_sha256":"522e6b9d517d9d871739bd4bdb47815952df9b5a44bc63886834e0f608fd936e","source":{"kind":"arxiv","id":"2606.09019","version":1},"attestation_state":"computed","paper":{"title":"TLDR: Compressing Audio Tokens for Efficient Autoregressive Text-to-Speech","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SD","authors_text":"Heeseung Kim, Hyoeun Kim, Hyunjin Choi, Junwon Moon, Kyuhong Shim, Yejin Lee","submitted_at":"2026-06-08T04:32:08Z","abstract_excerpt":"Codec-based autoregressive (AR) speech language models have achieved strong text-to-speech (TTS) quality by modeling speech as sequences of discrete audio tokens with large pretrained backbones. However, this token-level formulation creates a structural efficiency bottleneck: speech-token sequences are much longer than text sequences, requiring the AR backbone to perform causal computation at every token position and maintain a KV cache that grows with the sequence length. We introduce TLDR, a patch-based autoregressive framework that accelerates codec-based AR-TTS by shifting the causal model"},"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":"2606.09019","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SD","submitted_at":"2026-06-08T04:32:08Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f2c60da6622f65f7fc6c99f5542e70b987d310e4d70ff57f5e3852f050d88632","abstract_canon_sha256":"ba9795d1edfa4648d9cc3c85ff3b199adbf78cc82ffe5848135725bfe91cd4e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:07:53.870529Z","signature_b64":"QePmSbyjVMjmiV9DxBi1XMkrROk3/uITTKswIq1gVGrqHmfZA9hzYJeduW8m2FXbB6boro4KqBz6TqMjNq6yBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"522e6b9d517d9d871739bd4bdb47815952df9b5a44bc63886834e0f608fd936e","last_reissued_at":"2026-06-09T02:07:53.869620Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:07:53.869620Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TLDR: Compressing Audio Tokens for Efficient Autoregressive Text-to-Speech","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SD","authors_text":"Heeseung Kim, Hyoeun Kim, Hyunjin Choi, Junwon Moon, Kyuhong Shim, Yejin Lee","submitted_at":"2026-06-08T04:32:08Z","abstract_excerpt":"Codec-based autoregressive (AR) speech language models have achieved strong text-to-speech (TTS) quality by modeling speech as sequences of discrete audio tokens with large pretrained backbones. However, this token-level formulation creates a structural efficiency bottleneck: speech-token sequences are much longer than text sequences, requiring the AR backbone to perform causal computation at every token position and maintain a KV cache that grows with the sequence length. We introduce TLDR, a patch-based autoregressive framework that accelerates codec-based AR-TTS by shifting the causal model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09019","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/2606.09019/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":"2606.09019","created_at":"2026-06-09T02:07:53.869774+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.09019v1","created_at":"2026-06-09T02:07:53.869774+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09019","created_at":"2026-06-09T02:07:53.869774+00:00"},{"alias_kind":"pith_short_12","alias_value":"KIXGXHKRPWOY","created_at":"2026-06-09T02:07:53.869774+00:00"},{"alias_kind":"pith_short_16","alias_value":"KIXGXHKRPWOYOFZZ","created_at":"2026-06-09T02:07:53.869774+00:00"},{"alias_kind":"pith_short_8","alias_value":"KIXGXHKR","created_at":"2026-06-09T02:07:53.869774+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/KIXGXHKRPWOYOFZZXVF5WR4BLF","json":"https://pith.science/pith/KIXGXHKRPWOYOFZZXVF5WR4BLF.json","graph_json":"https://pith.science/api/pith-number/KIXGXHKRPWOYOFZZXVF5WR4BLF/graph.json","events_json":"https://pith.science/api/pith-number/KIXGXHKRPWOYOFZZXVF5WR4BLF/events.json","paper":"https://pith.science/paper/KIXGXHKR"},"agent_actions":{"view_html":"https://pith.science/pith/KIXGXHKRPWOYOFZZXVF5WR4BLF","download_json":"https://pith.science/pith/KIXGXHKRPWOYOFZZXVF5WR4BLF.json","view_paper":"https://pith.science/paper/KIXGXHKR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.09019&json=true","fetch_graph":"https://pith.science/api/pith-number/KIXGXHKRPWOYOFZZXVF5WR4BLF/graph.json","fetch_events":"https://pith.science/api/pith-number/KIXGXHKRPWOYOFZZXVF5WR4BLF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KIXGXHKRPWOYOFZZXVF5WR4BLF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KIXGXHKRPWOYOFZZXVF5WR4BLF/action/storage_attestation","attest_author":"https://pith.science/pith/KIXGXHKRPWOYOFZZXVF5WR4BLF/action/author_attestation","sign_citation":"https://pith.science/pith/KIXGXHKRPWOYOFZZXVF5WR4BLF/action/citation_signature","submit_replication":"https://pith.science/pith/KIXGXHKRPWOYOFZZXVF5WR4BLF/action/replication_record"}},"created_at":"2026-06-09T02:07:53.869774+00:00","updated_at":"2026-06-09T02:07:53.869774+00:00"}