{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BN25L3TV7SWS56NPPWA3ONGQTQ","short_pith_number":"pith:BN25L3TV","schema_version":"1.0","canonical_sha256":"0b75d5ee75fcad2ef9af7d81b734d09c1ad935185bf6b4ac2a77196aa118d6cc","source":{"kind":"arxiv","id":"2601.22661","version":2},"attestation_state":"computed","paper":{"title":"Evaluating and Rewarding LALMs for Expressive Role-Play TTS via Mean Continuation Log-Probability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Cheng Yi, Haiyang Sun, Hao Gu, Jingbei Li, Xuerui Yang, Ye Bai, Yechang Huang, Yong Ren, Yujie Chen","submitted_at":"2026-01-30T07:27:48Z","abstract_excerpt":"Recent advances in Large Audio Language Models (LALMs) have extended Text-to-Speech (TTS) to interactive role-play scenarios, which demand high expressiveness and strict adherence to role-play instructions. However, existing models struggle to maintain stylistic consistency with character profiles and scene descriptions across multi-turn dialogues. A critical bottleneck is the lack of objective metrics for quantifying speaking style. To bridge this gap, we propose Mean Continuation Log-Probability (MCLP) as both an evaluation metric and a reward signal, validated on LALM-based Role-Play TTS (R"},"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":"2601.22661","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-01-30T07:27:48Z","cross_cats_sorted":[],"title_canon_sha256":"59e40240fda2b75573c96ab4f4c1a52e87c3b6ee4361d35f778b9a14a90de795","abstract_canon_sha256":"d727c0ba2aded190a2f2f4236cb6cb2a8c5df3c80892d3f4001c41ccd353bdfd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:04:36.759070Z","signature_b64":"R4PjSiHvfl23ka2PillzuVNK7WSOoRRYRRF490v4FSh6pNxxQc2ywQZ19XeBlizkxtX7DPBXeNAlDHKDz3CSAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b75d5ee75fcad2ef9af7d81b734d09c1ad935185bf6b4ac2a77196aa118d6cc","last_reissued_at":"2026-05-29T01:04:36.758521Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:04:36.758521Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating and Rewarding LALMs for Expressive Role-Play TTS via Mean Continuation Log-Probability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SD","authors_text":"Cheng Yi, Haiyang Sun, Hao Gu, Jingbei Li, Xuerui Yang, Ye Bai, Yechang Huang, Yong Ren, Yujie Chen","submitted_at":"2026-01-30T07:27:48Z","abstract_excerpt":"Recent advances in Large Audio Language Models (LALMs) have extended Text-to-Speech (TTS) to interactive role-play scenarios, which demand high expressiveness and strict adherence to role-play instructions. However, existing models struggle to maintain stylistic consistency with character profiles and scene descriptions across multi-turn dialogues. A critical bottleneck is the lack of objective metrics for quantifying speaking style. To bridge this gap, we propose Mean Continuation Log-Probability (MCLP) as both an evaluation metric and a reward signal, validated on LALM-based Role-Play TTS (R"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.22661","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/2601.22661/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":"2601.22661","created_at":"2026-05-29T01:04:36.758592+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.22661v2","created_at":"2026-05-29T01:04:36.758592+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.22661","created_at":"2026-05-29T01:04:36.758592+00:00"},{"alias_kind":"pith_short_12","alias_value":"BN25L3TV7SWS","created_at":"2026-05-29T01:04:36.758592+00:00"},{"alias_kind":"pith_short_16","alias_value":"BN25L3TV7SWS56NP","created_at":"2026-05-29T01:04:36.758592+00:00"},{"alias_kind":"pith_short_8","alias_value":"BN25L3TV","created_at":"2026-05-29T01:04:36.758592+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/BN25L3TV7SWS56NPPWA3ONGQTQ","json":"https://pith.science/pith/BN25L3TV7SWS56NPPWA3ONGQTQ.json","graph_json":"https://pith.science/api/pith-number/BN25L3TV7SWS56NPPWA3ONGQTQ/graph.json","events_json":"https://pith.science/api/pith-number/BN25L3TV7SWS56NPPWA3ONGQTQ/events.json","paper":"https://pith.science/paper/BN25L3TV"},"agent_actions":{"view_html":"https://pith.science/pith/BN25L3TV7SWS56NPPWA3ONGQTQ","download_json":"https://pith.science/pith/BN25L3TV7SWS56NPPWA3ONGQTQ.json","view_paper":"https://pith.science/paper/BN25L3TV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.22661&json=true","fetch_graph":"https://pith.science/api/pith-number/BN25L3TV7SWS56NPPWA3ONGQTQ/graph.json","fetch_events":"https://pith.science/api/pith-number/BN25L3TV7SWS56NPPWA3ONGQTQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BN25L3TV7SWS56NPPWA3ONGQTQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BN25L3TV7SWS56NPPWA3ONGQTQ/action/storage_attestation","attest_author":"https://pith.science/pith/BN25L3TV7SWS56NPPWA3ONGQTQ/action/author_attestation","sign_citation":"https://pith.science/pith/BN25L3TV7SWS56NPPWA3ONGQTQ/action/citation_signature","submit_replication":"https://pith.science/pith/BN25L3TV7SWS56NPPWA3ONGQTQ/action/replication_record"}},"created_at":"2026-05-29T01:04:36.758592+00:00","updated_at":"2026-05-29T01:04:36.758592+00:00"}