Abstract
Long-form subtitle translation requires reasoning over discourse and cultural context spanning episodes or entire series, while maintaining consistent terminology and style. Existing single-LLM methods are largely sentence-level, and multi-agent systems often use static workflows that do not adapt to scene complexity or production context. We propose SMART, a Self-evolving Multi-Agent system for long-foRm subtitle Translation. During test-time training, SMART builds persistent series-level memory and translates a subset of sentences through a dynamic router and Mixture-of-Agents layer with tools for terminology verification, subtitle constraint validation, and contextual retrieval. A judge-refiner loop scores candidates and uses textual critiques to update agent prompts and routing policies without retraining the underlying LLMs. During test-time inference, the evolved configuration translates the remaining series. We also introduce Subtitle Arena, covering 14 genres, 2--198 episodes per series, production years 1959--2023, and 15 target locales, together with SubMQM, a subtitle-adapted MQM framework with seven dimensions and 19 error categories. SMART achieves the best overall MQM score in all 15 Subtitle Arena directions, reducing average penalty by 6.9% over the strongest competing agent system. On a public benchmark, MuSC, SMART obtains the best model result across all 4 language pairs. SMART also achieves the best result in human evaluation with an overall score of 4.50/5.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Jin, H., Li, X., Sadoughi, N., Liu, Y., Wang, Y., Liu, Z., & Liu, Y. (2026). Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle Translation. https://omanscience.com/en/articles/breaking-babel-a-self-evolving-multi-agent-system-for-long-form-subtitle-translation
MLA 9
Jin, Haibo, et al. "Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle Translation." https://omanscience.com/en/articles/breaking-babel-a-self-evolving-multi-agent-system-for-long-form-subtitle-translation.
Chicago (author–date)
Jin, Haibo, Xinjie Li, Najmeh Sadoughi, Yang Liu, Yibo Wang, Zhu Liu, and Yuzong Liu. 2026. "Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle Translation." https://omanscience.com/en/articles/breaking-babel-a-self-evolving-multi-agent-system-for-long-form-subtitle-translation.
Harvard
Jin, H., Li, X., Sadoughi, N., Liu, Y., Wang, Y., Liu, Z. and Liu, Y. (2026) 'Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle Translation', Available at: https://omanscience.com/en/articles/breaking-babel-a-self-evolving-multi-agent-system-for-long-form-subtitle-translation.
Vancouver
Jin H, Li X, Sadoughi N, Liu Y, Wang Y, Liu Z, et al. Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle Translation. https://omanscience.com/en/articles/breaking-babel-a-self-evolving-multi-agent-system-for-long-form-subtitle-translation
IEEE
H. Jin, X. Li, N. Sadoughi, Y. Liu, Y. Wang, Z. Liu, and Y. Liu, "Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle Translation," https://omanscience.com/en/articles/breaking-babel-a-self-evolving-multi-agent-system-for-long-form-subtitle-translation.