{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WBPR5L5G5TIW7POMPZV4K4F5YW","short_pith_number":"pith:WBPR5L5G","schema_version":"1.0","canonical_sha256":"b05f1eafa6ecd16fbdcc7e6bc570bdc5ad4b20bd23024192eea55f17e82a9624","source":{"kind":"arxiv","id":"2509.11084","version":1},"attestation_state":"computed","paper":{"title":"Length-Aware Rotary Position Embedding for Text-Speech Alignment","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Hyeongju Kim, Jacob Morton, Jinhyeok Yang, Juheon Lee","submitted_at":"2025-09-14T04:25:13Z","abstract_excerpt":"Many recent text-to-speech (TTS) systems are built on transformer architectures and employ cross-attention mechanisms for text-speech alignment. Within these systems, rotary position embedding (RoPE) is commonly used to encode positional information in text and speech representations. In this work, we introduce length-aware RoPE (LARoPE), a simple yet effective extension of RoPE that improves text-speech alignment. Unlike RoPE, which relies on absolute indices, LARoPE computes relative distances between query and key positions using length-normalized indices. Experimental results show that LAR"},"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":"2509.11084","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"eess.AS","submitted_at":"2025-09-14T04:25:13Z","cross_cats_sorted":["cs.AI","cs.CL","cs.SD"],"title_canon_sha256":"8dcf7df60a38ef1c4f1374483e612ccbe4f56d6bab174b984cfd939172edd5b6","abstract_canon_sha256":"94510fb12597ebdb80b4beb57194e91a5f88574dd7a2337c66df3bd8b74d5f12"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:33.246630Z","signature_b64":"CJt7y1/A/zcrnyIU0CzGbuRPsW1eCmTX55edMccn7YCvyHacJnKDy0NE12k55GBgJZY6PgxehVOC0L/2/ffTAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b05f1eafa6ecd16fbdcc7e6bc570bdc5ad4b20bd23024192eea55f17e82a9624","last_reissued_at":"2026-07-05T12:11:33.246116Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:33.246116Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Length-Aware Rotary Position Embedding for Text-Speech Alignment","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.SD"],"primary_cat":"eess.AS","authors_text":"Hyeongju Kim, Jacob Morton, Jinhyeok Yang, Juheon Lee","submitted_at":"2025-09-14T04:25:13Z","abstract_excerpt":"Many recent text-to-speech (TTS) systems are built on transformer architectures and employ cross-attention mechanisms for text-speech alignment. Within these systems, rotary position embedding (RoPE) is commonly used to encode positional information in text and speech representations. In this work, we introduce length-aware RoPE (LARoPE), a simple yet effective extension of RoPE that improves text-speech alignment. Unlike RoPE, which relies on absolute indices, LARoPE computes relative distances between query and key positions using length-normalized indices. Experimental results show that LAR"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.11084","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/2509.11084/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":"2509.11084","created_at":"2026-07-05T12:11:33.246181+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.11084v1","created_at":"2026-07-05T12:11:33.246181+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.11084","created_at":"2026-07-05T12:11:33.246181+00:00"},{"alias_kind":"pith_short_12","alias_value":"WBPR5L5G5TIW","created_at":"2026-07-05T12:11:33.246181+00:00"},{"alias_kind":"pith_short_16","alias_value":"WBPR5L5G5TIW7POM","created_at":"2026-07-05T12:11:33.246181+00:00"},{"alias_kind":"pith_short_8","alias_value":"WBPR5L5G","created_at":"2026-07-05T12:11:33.246181+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.22083","citing_title":"RobustSpeechFlow: Learning Robust Text-to-Speech Trajectories via Augmentation-based Contrastive Flow Matching","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW","json":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW.json","graph_json":"https://pith.science/api/pith-number/WBPR5L5G5TIW7POMPZV4K4F5YW/graph.json","events_json":"https://pith.science/api/pith-number/WBPR5L5G5TIW7POMPZV4K4F5YW/events.json","paper":"https://pith.science/paper/WBPR5L5G"},"agent_actions":{"view_html":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW","download_json":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW.json","view_paper":"https://pith.science/paper/WBPR5L5G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.11084&json=true","fetch_graph":"https://pith.science/api/pith-number/WBPR5L5G5TIW7POMPZV4K4F5YW/graph.json","fetch_events":"https://pith.science/api/pith-number/WBPR5L5G5TIW7POMPZV4K4F5YW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW/action/storage_attestation","attest_author":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW/action/author_attestation","sign_citation":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW/action/citation_signature","submit_replication":"https://pith.science/pith/WBPR5L5G5TIW7POMPZV4K4F5YW/action/replication_record"}},"created_at":"2026-07-05T12:11:33.246181+00:00","updated_at":"2026-07-05T12:11:33.246181+00:00"}