{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TNGIYICMVWZGTNQKA427AVMEZC","short_pith_number":"pith:TNGIYICM","canonical_record":{"source":{"id":"2507.05911","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-07-08T11:57:16Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"6d804c488979d11f55dc073fc4d9ea9e37cfbf8ffa4733a90607bee48debd178","abstract_canon_sha256":"2ededa7b24c991c49c6cb1008c684e3b7ef3b54fdce13e0639fca509678cb350"},"schema_version":"1.0"},"canonical_sha256":"9b4c8c204cadb269b60a0735f05584c8b6de253c14b85a9c0e2a90b9f5b13454","source":{"kind":"arxiv","id":"2507.05911","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.05911","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"arxiv_version","alias_value":"2507.05911v1","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.05911","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"pith_short_12","alias_value":"TNGIYICMVWZG","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"pith_short_16","alias_value":"TNGIYICMVWZGTNQK","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"pith_short_8","alias_value":"TNGIYICM","created_at":"2026-07-05T11:33:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TNGIYICMVWZGTNQKA427AVMEZC","target":"record","payload":{"canonical_record":{"source":{"id":"2507.05911","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-07-08T11:57:16Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"6d804c488979d11f55dc073fc4d9ea9e37cfbf8ffa4733a90607bee48debd178","abstract_canon_sha256":"2ededa7b24c991c49c6cb1008c684e3b7ef3b54fdce13e0639fca509678cb350"},"schema_version":"1.0"},"canonical_sha256":"9b4c8c204cadb269b60a0735f05584c8b6de253c14b85a9c0e2a90b9f5b13454","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:38.264667Z","signature_b64":"E/dtw9bIWQm8t3nliIOGgXnFtQ2WDRuO2PmmNhvgLAhSlA+R3AzLi1GVvru69qF1r2LMdSn8Qmb21ioJBG34Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9b4c8c204cadb269b60a0735f05584c8b6de253c14b85a9c0e2a90b9f5b13454","last_reissued_at":"2026-07-05T11:33:38.264092Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:38.264092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.05911","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:33:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4iCou4UTD6yfvHnuDgavFZVbXW+sA/1fs6g4Ld4tstttIJp47TJXRVCeg5jWn1Pgc02N+GdqLAqBETFcu1FyAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T07:19:17.040823Z"},"content_sha256":"c31281d3aad6749fbbf79e4a975c3e9d9cf0bd19e22e4ac578b7ce0e0533d0ac","schema_version":"1.0","event_id":"sha256:c31281d3aad6749fbbf79e4a975c3e9d9cf0bd19e22e4ac578b7ce0e0533d0ac"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TNGIYICMVWZGTNQKA427AVMEZC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Differentiable Reward Optimization for LLM based TTS system","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Changfeng Gao, Shiliang Zhang, Zhihao Du","submitted_at":"2025-07-08T11:57:16Z","abstract_excerpt":"This paper proposes a novel Differentiable Reward Optimization (DiffRO) method aimed at enhancing the performance of neural codec language models based text-to-speech (TTS) systems. In contrast to conventional reinforcement learning from human feedback (RLHF) approaches applied to TTS, DiffRO directly compute the rewards based on neural codec tokens, rather than relying on synthesized audio. Furthermore, we employ the Gumbel-Softmax technique to render the reward function differentiable, thereby streamlining the RLHF training process. Additionally, we introduce a multi-task reward (MTR) model "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.05911","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/2507.05911/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:33:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"02JlxGg8VUtv/gpqIHBRl3Cw+Z0Juw2RPA+U0vTJm+OFFrViRWqetEsLSJ1YnDof3pRlO90K7vs4i63K/KiHCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T07:19:17.041392Z"},"content_sha256":"4934713464d9c69fb0e224f0dc9a530df817194d73261ae278949493670a422d","schema_version":"1.0","event_id":"sha256:4934713464d9c69fb0e224f0dc9a530df817194d73261ae278949493670a422d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TNGIYICMVWZGTNQKA427AVMEZC/bundle.json","state_url":"https://pith.science/pith/TNGIYICMVWZGTNQKA427AVMEZC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TNGIYICMVWZGTNQKA427AVMEZC/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-07T07:19:17Z","links":{"resolver":"https://pith.science/pith/TNGIYICMVWZGTNQKA427AVMEZC","bundle":"https://pith.science/pith/TNGIYICMVWZGTNQKA427AVMEZC/bundle.json","state":"https://pith.science/pith/TNGIYICMVWZGTNQKA427AVMEZC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TNGIYICMVWZGTNQKA427AVMEZC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TNGIYICMVWZGTNQKA427AVMEZC","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2ededa7b24c991c49c6cb1008c684e3b7ef3b54fdce13e0639fca509678cb350","cross_cats_sorted":["cs.AI","eess.AS"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-07-08T11:57:16Z","title_canon_sha256":"6d804c488979d11f55dc073fc4d9ea9e37cfbf8ffa4733a90607bee48debd178"},"schema_version":"1.0","source":{"id":"2507.05911","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.05911","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"arxiv_version","alias_value":"2507.05911v1","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.05911","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"pith_short_12","alias_value":"TNGIYICMVWZG","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"pith_short_16","alias_value":"TNGIYICMVWZGTNQK","created_at":"2026-07-05T11:33:38Z"},{"alias_kind":"pith_short_8","alias_value":"TNGIYICM","created_at":"2026-07-05T11:33:38Z"}],"graph_snapshots":[{"event_id":"sha256:4934713464d9c69fb0e224f0dc9a530df817194d73261ae278949493670a422d","target":"graph","created_at":"2026-07-05T11:33:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2507.05911/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper proposes a novel Differentiable Reward Optimization (DiffRO) method aimed at enhancing the performance of neural codec language models based text-to-speech (TTS) systems. In contrast to conventional reinforcement learning from human feedback (RLHF) approaches applied to TTS, DiffRO directly compute the rewards based on neural codec tokens, rather than relying on synthesized audio. Furthermore, we employ the Gumbel-Softmax technique to render the reward function differentiable, thereby streamlining the RLHF training process. Additionally, we introduce a multi-task reward (MTR) model ","authors_text":"Changfeng Gao, Shiliang Zhang, Zhihao Du","cross_cats":["cs.AI","eess.AS"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-07-08T11:57:16Z","title":"Differentiable Reward Optimization for LLM based TTS system"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.05911","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c31281d3aad6749fbbf79e4a975c3e9d9cf0bd19e22e4ac578b7ce0e0533d0ac","target":"record","created_at":"2026-07-05T11:33:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2ededa7b24c991c49c6cb1008c684e3b7ef3b54fdce13e0639fca509678cb350","cross_cats_sorted":["cs.AI","eess.AS"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-07-08T11:57:16Z","title_canon_sha256":"6d804c488979d11f55dc073fc4d9ea9e37cfbf8ffa4733a90607bee48debd178"},"schema_version":"1.0","source":{"id":"2507.05911","kind":"arxiv","version":1}},"canonical_sha256":"9b4c8c204cadb269b60a0735f05584c8b6de253c14b85a9c0e2a90b9f5b13454","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9b4c8c204cadb269b60a0735f05584c8b6de253c14b85a9c0e2a90b9f5b13454","first_computed_at":"2026-07-05T11:33:38.264092Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:33:38.264092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"E/dtw9bIWQm8t3nliIOGgXnFtQ2WDRuO2PmmNhvgLAhSlA+R3AzLi1GVvru69qF1r2LMdSn8Qmb21ioJBG34Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:33:38.264667Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.05911","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c31281d3aad6749fbbf79e4a975c3e9d9cf0bd19e22e4ac578b7ce0e0533d0ac","sha256:4934713464d9c69fb0e224f0dc9a530df817194d73261ae278949493670a422d"],"state_sha256":"d93e8b5afd25ec16595822a4761e150a35afcb63ce62695fdf8ccbbdac1c80bb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JH19C/j1Og5Bio/hjHMgyg5km1g18w86I1laRPuXna32Cgds/2+EI92PdoJzFFq2L1DWHES+KFffXCsF8bkIDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T07:19:17.049437Z","bundle_sha256":"e716a353c4554a746b98ae2ec4cf375b24d3df5e8f2646641ebeca34d3a4d770"}}