DPO-Tuned Large Language Models for Segmentation in Simultaneous Speech Translation

• URL: http://arxiv.org/abs/2510.12195v1
• Date: Tue, 14 Oct 2025 06:41:36 GMT
• Title: DPO-Tuned Large Language Models for Segmentation in Simultaneous Speech Translation
• Authors: Zeyu Yang, Satoshi Nakamura,
• Abstract summary: Simultaneous speech translation requires accurate segmentation to balance translation quality and latency.<n>We propose a segmentation framework based on large language models trained with Direct Preference Optimization (DPO)<n>By leveraging preference alignment, our method enables LLMs to predict natural segmentation points that better meet the demands of real-time translation.
• Score: 6.611635315225665
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Simultaneous speech translation requires accurate segmentation to balance translation quality and latency. Recent studies such as SHAS have introduced pretrained segmentation models, achieving stronger performance than heuristic rules. However, segmentation models such as SHAS, though pretrained and more robust than heuristic methods, are still constrained by supervised learning objectives and do not incorporate human preference alignment, which is crucial for natural real-time interpretation. In this work, we propose a segmentation framework based on large language models (LLMs) trained with Direct Preference Optimization (DPO). By leveraging preference alignment, our method enables LLMs to predict natural segmentation points that better meet the demands of real-time translation. We evaluate the system on the ACL 60/60 corpus across three language pairs (English-Japanese, Chinese, German), using SeamlessM4T v2 as the translation backbone. Experimental results show that our DPO-tuned LLM achieves higher segmentation accuracy than SHAS and yields consistent improvements in translation quality (BLEU, COMET) as well as latency (Average Lagging). Furthermore, our system benefits from IWSLT baselines for direct comparison. These findings highlight the potential of preference-tuned LLMs to surpass existing pretrained segmentation models and advance adaptive, human-aligned simultaneous interpretation.

Related papers

• Beyond Many-Shot Translation: Scaling In-Context Demonstrations For Low-Resource Machine Translation [49.82863380286994]
  In-context learning may offer novel ways to adapt Large Language Models for low-resource machine translation.<n>In this study, we explore scaling low-resource machine translation ICL beyond the few-shot setting to thousands of examples with long-context models.<n>Our experiments on Javanese and Sundanese show that gains from additional context saturate quickly and can degrade near the maximum context window.
  arXiv  Detail & Related papers  (2026-02-04T17:02:22Z)
• From Utterance to Vividity: Training Expressive Subtitle Translation LLM via Adaptive Local Preference Optimization [12.547838537411215]
  We focus on how to construct translation LLMs that meet the needs of domain customization.<n>We take visual media subtitle translation as our topic and explore how to train expressive and vivid translation LLMs.
  arXiv  Detail & Related papers  (2026-02-01T07:24:06Z)
• MLLP-VRAIN UPV system for the IWSLT 2025 Simultaneous Speech Translation Translation task [7.247809853198223]
  This work describes the participation of the MLLP-VRAIN research group in the shared task of the IWSLT 2025 Simultaneous Speech Translation track.<n>Our submission addresses the unique challenges of real-time translation of long-form speech by developing a modular cascade system.
  arXiv  Detail & Related papers  (2025-06-23T16:44:01Z)
• Trans-Zero: Self-Play Incentivizes Large Language Models for Multilingual Translation Without Parallel Data [64.4458540273004]
  We propose a self-play framework that leverages only monolingual data and the intrinsic multilingual knowledge of Large Language Models (LLMs)<n>Experiments demonstrate that this approach not only matches the performance of models trained on large-scale parallel data but also excels in non-English translation directions.
  arXiv  Detail & Related papers  (2025-04-20T16:20:30Z)
• Latent Thought Models with Variational Bayes Inference-Time Computation [52.63299874322121]
  Latent Thought Models (LTMs) incorporate explicit latent thought vectors that follow an explicit prior model in latent space.<n>LTMs demonstrate superior sample and parameter efficiency compared to autoregressive models and discrete diffusion models.
  arXiv  Detail & Related papers  (2025-02-03T17:50:34Z)
• Preference Alignment Improves Language Model-Based TTS [76.70693823683091]
  preference alignment algorithms adjust LMs to align with the preferences of reward models, enhancing the desirability of the generated content. With a 1.15B parameter LM-based TTS model, we demonstrate that preference alignment consistently improves intelligibility, speaker similarity, and proxy subjective evaluation scores.
  arXiv  Detail & Related papers  (2024-09-19T01:58:19Z)
• Self-Exploring Language Models: Active Preference Elicitation for Online Alignment [88.56809269990625]
  We propose a bilevel objective optimistically biased towards potentially high-reward responses to actively explore out-of-distribution regions. Our experimental results demonstrate that when fine-tuned on Zephyr-7B-SFT and Llama-3-8B-Instruct models, Self-Exploring Language Models (SELM) significantly boosts the performance on instruction-following benchmarks.
  arXiv  Detail & Related papers  (2024-05-29T17:59:07Z)
• A Preference-driven Paradigm for Enhanced Translation with Large Language Models [33.51585908894444]
  Large language models (LLMs) can achieve remarkable translation performance using only a small amount of parallel data. SFT simply instructs the model to imitate the reference translations at the token level, making it vulnerable to the noise present in the references. We propose a preference-based approach built upon the Plackett-Luce model to overcome this plateau.
  arXiv  Detail & Related papers  (2024-04-17T11:52:47Z)
• Building Accurate Translation-Tailored LLMs with Language Aware Instruction Tuning [57.323716555996114]
  Off-target translation remains an unsolved problem, especially for low-resource languages. Recent works have either designed advanced prompting strategies to highlight the functionality of translation instructions or exploited the in-context learning ability of LLMs. In this work, we design a two-stage fine-tuning algorithm to improve the instruction-following ability (especially the translation direction) of LLMs.
  arXiv  Detail & Related papers  (2024-03-21T13:47:40Z)
• Tokenizer Choice For LLM Training: Negligible or Crucial? [30.33170936148845]
  We study the influence of tokenizer choice on Large Language Models (LLMs) downstream performance by training 24 mono- and multilingual LLMs. We find that the tokenizer choice can significantly impact the model's downstream performance and training costs. We show that multilingual tokenizers trained on the five most frequent European languages require vocabulary size increases of factor three in comparison to English.
  arXiv  Detail & Related papers  (2023-10-12T22:44:19Z)
• Improving Multilingual Translation by Representation and Gradient Regularization [82.42760103045083]
  We propose a joint approach to regularize NMT models at both representation-level and gradient-level. Our results demonstrate that our approach is highly effective in both reducing off-target translation occurrences and improving zero-shot translation performance.
  arXiv  Detail & Related papers  (2021-09-10T10:52:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.