From Utterance to Vividity: Training Expressive Subtitle Translation LLM via Adaptive Local Preference Optimization
This addresses the need for domain-customized translation in visual media subtitles, representing an incremental improvement over existing LLM methods.
The study tackled the problem of training expressive and vivid translation LLMs for visual media subtitles by constructing a multidirectional subtitle parallel corpus and proposing the Adaptive Local Preference Optimization (ALPO) method, achieving outstanding performance in multidimensional translation quality evaluation.
The rapid development of Large Language Models (LLMs) has significantly enhanced the general capabilities of machine translation. However, as application scenarios become more complex, the limitations of LLMs in vertical domain translations are gradually becoming apparent. In this study, we focus on how to construct translation LLMs that meet the needs of domain customization. We take visual media subtitle translation as our topic and explore how to train expressive and vivid translation LLMs. We investigated the situations of subtitle translation and other domains of literal and liberal translation, verifying the reliability of LLM as reward model and evaluator for translation. Additionally, to train an expressive translation LLM, we constructed and released a multidirectional subtitle parallel corpus dataset and proposed the Adaptive Local Preference Optimization (ALPO) method to address fine-grained preference alignment. Experimental results demonstrate that ALPO achieves outstanding performance in multidimensional evaluation of translation quality.