Paper Citation Record · LEDGER

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech

As of 14 August 2026, this Paper Citation Record lists 37 of 37 outbound references and 2 inbound Pith citation observations for arXiv:2509.00685.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2509.00685 v1

Coverage vector

measured 37 of 37 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-05T13:24:34.291292Z

measured 39 of 39 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-14T06:32:32.682623+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-05T13:24:34.127489Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-07-03T23:39:03.102169Z

Reference resolution

37 of 37 outbound references displayed

verified exact1
verified fuzzy20
unresolved16
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation c1474778-faef-41c8-8129-2ccfac66fa38 · outbound

This paper cites LM-based TTS systems convert speech waveforms into sequences of discrete tokens using neural audio codecs [1, 2, 3, 4, 5] and operate in a discrete space [6, 7].

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech LM-based TTS systems convert speech waveforms into sequences of discrete tokens using neural audio codecs [1, 2, 3, 4, 5] and operate in a discrete space [6, 7]

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.888239Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 38e3ca5f-8b4e-4350-8104-c02830c7d831 · outbound

This paper cites Preference Alignment Preference alignment is often formatted as a reinforcement learning problem.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Preference Alignment Preference alignment is often formatted as a reinforcement learning problem

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.872374Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 96266f2d-f5fc-4393-93df-12378db3151d · outbound

This paper cites MPO involves constructing a multidi- mensional preference dataset and incorporating additional reg- ularization during training to prevent model degradation.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech MPO involves constructing a multidi- mensional preference dataset and incorporating additional reg- ularization during training to prevent model degradation

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.857606Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 8a2a5532-4b7e-4db1-b97d-9796034dec0d · outbound

This paper cites an unresolved cited work.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Unresolved cited work

Reference 4

Resolution: unresolved; raw_fallback, observed 2026-08-05T13:24:34.843343Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 4ef14480-6bc2-4ddc-bf8e-f8dc453829ba · outbound

This paper cites an unresolved cited work.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Unresolved cited work

Reference 5

Resolution: unresolved; raw_fallback, observed 2026-08-05T13:24:34.801594Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 801c9947-2627-41c8-acd0-93c70e48918f · outbound

This paper cites Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.185099Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d77f167f-eacc-4b15-befe-d7600d999574 · outbound

This paper cites The Interspeech 2024 Challenge on Speech Processing Using Discrete Units.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech The Interspeech 2024 Challenge on Speech Processing Using Discrete Units

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-08-05T13:24:34.500670Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation cc4a7448-e733-4358-ac93-788c0a435def · outbound

This paper cites BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech BASE TTS: Lessons from building a billion-parameter Text-to-Speech model on 100K hours of data

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.194579Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7474ce0d-c671-43e8-ba9e-1894bebeb606 · outbound

This paper cites Soundstream: An end-to-end neural audio codec,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Soundstream: An end-to-end neural audio codec,

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.787102Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation d467e4ce-079c-4266-8cfc-768027922cbc · outbound

This paper cites High fidelity neural audio compression,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech High fidelity neural audio compression,

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.772980Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation a34a4908-4f10-4779-af63-17074227dd82 · outbound

This paper cites Funcodec: A funda- mental, reproducible and integrable open-source toolkit for neural speech codec,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Funcodec: A funda- mental, reproducible and integrable open-source toolkit for neural speech codec,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.758699Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation aef01163-c347-47e7-98ce-39d5fe2a0876 · outbound

This paper cites Speechtok- enizer: Unified speech tokenizer for speech language models,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Speechtok- enizer: Unified speech tokenizer for speech language models,

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.740531Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation e8a331b1-aeb0-4d71-9d8b-05979977370d · outbound

This paper cites Single-codec: Single-codebook speech codec towards high-performance speech generation,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Single-codec: Single-codebook speech codec towards high-performance speech generation,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.725039Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 09c8a629-f739-4481-b3c8-32a1a3d5b399 · outbound

This paper cites GPT-4 Technical Report.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech GPT-4 Technical Report

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.223367Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 83ec8996-46a1-4a65-8e85-d84b6076b1aa · outbound

This paper cites Simpo: Simple preference opti- mization with a reference-free reward,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Simpo: Simple preference opti- mization with a reference-free reward,

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.664942Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation a4029945-90a5-4bdd-bf1a-6fef626d8fd9 · outbound

This paper cites MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.127489Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5e241448-ffe0-4bc6-a5e5-a40fb9a69705 · outbound

This paper cites V oice- craft: Zero-shot speech editing and text-to-speech in the wild,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech V oice- craft: Zero-shot speech editing and text-to-speech in the wild,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.709223Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation f8d55cdd-b927-4d12-b893-ac9fee5a77e7 · outbound

This paper cites Seed-TTS: A Family of High-Quality Versatile Speech Generation Models.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Seed-TTS: A Family of High-Quality Versatile Speech Generation Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.204193Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9d6e5a8-229c-4010-ab08-c9020fe60d4e · outbound

This paper cites Llasa: Scaling Train-Time and Inference-Time Compute for Llama-based Speech Synthesis.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Llasa: Scaling Train-Time and Inference-Time Compute for Llama-based Speech Synthesis

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.208839Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ccaec03-6510-4a97-9f6b-eec943eeb06d · outbound

This paper cites Training language models to follow instructions with human feedback,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Training language models to follow instructions with human feedback,

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.694020Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 605e7fe9-3802-4c3f-a11e-f5307488ef54 · outbound

This paper cites Model alignment as prospect theoretic optimization,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Model alignment as prospect theoretic optimization,

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.679116Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation dbc466ba-c57f-495e-83fb-6dc2ed89498c · outbound

This paper cites Rank analysis of incomplete block designs: I. the method of paired comparisons,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Rank analysis of incomplete block designs: I. the method of paired comparisons,

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.259985Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c5981849-b668-4319-8159-8d5f58ab9545 · outbound

This paper cites Wenetspeech4tts: A 12,800-hour mandarin tts corpus for large speech generation model bench- mark,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Wenetspeech4tts: A 12,800-hour mandarin tts corpus for large speech generation model bench- mark,

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.611695Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation c8dba464-44dd-4ca3-a11e-fea4c77780e1 · outbound

This paper cites Direct preference optimization: Your language model is secretly a reward model,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Direct preference optimization: Your language model is secretly a reward model,

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.648710Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation a986b73c-56d6-4262-83eb-83801e5a483d · outbound

This paper cites This fine-tuned model serves as the baseline for our experiments.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech This fine-tuned model serves as the baseline for our experiments

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.829916Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation d36b11b8-b95d-4739-8938-71b33d34de74 · outbound

This paper cites Speechalign: Aligning speech generation to human preferences,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Speechalign: Aligning speech generation to human preferences,

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.237058Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4cf35e4-5319-4912-bffc-3276d6180c0d · outbound

This paper cites Enhancing Zero-shot Text-to-Speech Synthesis with Human Feedback.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Enhancing Zero-shot Text-to-Speech Synthesis with Human Feedback

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.241314Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 35f7e82a-5545-42ae-8daf-d8ba21770135 · outbound

This paper cites Robust Zero-Shot Text-to-Speech Synthesis with Reverse Inference Optimization.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Robust Zero-Shot Text-to-Speech Synthesis with Reverse Inference Optimization

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.245816Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d9a4cfd-7c26-4547-84b5-c8a37e367c2a · outbound

This paper cites Dynamic time warping is employed to align the generated and reference speech features of different sequential lengths, following the evaluation script in ESPnet [29].

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Dynamic time warping is employed to align the generated and reference speech features of different sequential lengths, following the evaluation script in ESPnet [29]

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.816105Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation ecaaff75-c909-468a-8a05-64daf778a292 · outbound

This paper cites Preference Alignment Improves Language Model-Based TTS.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Preference Alignment Improves Language Model-Based TTS

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.250279Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8134630d-ed0f-48c1-b116-42f1f1a20d78 · outbound

This paper cites Robust Preference Optimization through Reward Model Distillation.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Robust Preference Optimization through Reward Model Distillation

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.255346Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f67ff59-a87c-4426-aeda-8256a16813fe · outbound

This paper cites Libriheavy: A 50, 000 hours ASR corpus with punctuation casing and context,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Libriheavy: A 50, 000 hours ASR corpus with punctuation casing and context,

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.595842Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 26dd2f97-3dae-4597-922b-cb06a62ab168 · outbound

This paper cites The ISCSLP 2024 con- versational voice clone (covoc) challenge: Tasks, results and find- ings,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech The ISCSLP 2024 con- versational voice clone (covoc) challenge: Tasks, results and find- ings,

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.580999Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 8aa513c0-2240-4aff-a744-0d9fe6ed5c15 · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech LLaMA: Open and Efficient Foundation Language Models

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.278020Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9bfd8375-19f8-4303-a044-ce387eba3bae · outbound

This paper cites Paraformer: Fast and accurate parallel transformer for non-autoregressive end-to- end speech recognition,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Paraformer: Fast and accurate parallel transformer for non-autoregressive end-to- end speech recognition,

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.565726Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 16509b4a-4dd7-4961-a565-588a3a90e487 · outbound

This paper cites Large-scale self-supervised speech representation learning for automatic speaker verification,.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech Large-scale self-supervised speech representation learning for automatic speaker verification,

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T13:24:34.550700Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation aaa69b3d-f608-476b-b123-6fa3bc4dbe28 · outbound

This paper cites ESPnet2-TTS: Extending the Edge of TTS Research.

MPO: Multidimensional Preference Optimization for Language Model-based Text-to-Speech ESPnet2-TTS: Extending the Edge of TTS Research

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-05T13:24:34.291292Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation a4029945-90a5-4bdd-bf1a-6fef626d8fd9 · inbound