EvoID：面向身份保持视频生成的强化演化
EvoID: Reinforced Evolution for Identity-Preserving Video Generation
把身份保持视频生成从静态模仿学习改写为强化学习自演化过程，用「客观指标 + MLLM 整体质量评估」构成双路动态奖励，并用冻结教师锚定学生模型；做视频生成后训练或奖励设计的工作可直接借鉴这套奖励组合与教师锚定策略。
原文依据 a dynamic, dual-path reward mechanism, which acts as an intrinsic critic by adaptively combining objective metric indicators and MLLM-based holistic quality assessment
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出 EvoID，一种将身份保持视频生成重新表述为通过强化学习实现的自演化过程的新框架。EvoID 超越了模仿学习的静态范式，使生成模型能够主动学习并优化身份保真度、运动自然性与时序一致性之间的复杂权衡。EvoID 的核心是一种动态双路奖励机制，它通过自适应地结合客观指标与基于多模态大模型的整体质量评估，充当内在的评判者。这使得模型能够「演化」其生成策略，在训练的不同阶段聚焦于质量的不同方面。为确保稳定且连贯的演化，我们用冻结的教师模型锚定处于探索中的学生模型，在保留稳健世界先验的同时，允许在生成视频时进行创造性精修。大量实验证明了本方案的优越性：EvoID 在 OpenS2V-Eval 数据集的 Human-Domain 上取得 0.687 的总分，超过开源 VACE 的 0.658 与商业模型 Hailuo 的 0.653。此外，在我们新构建的、以人类感知为优先并更全面反映视频质量的基于多模态大模型的指标上，EvoID 还取得了 0.718 的新纪录。
英文原文摘要 1295 字符
We present EvoID, a novel framework that reformulates Identity-Preserving Video Generation as a self-evolving process through Reinforcement Learning. Moving beyond the static paradigm of imitation learning, EvoID enables a generative model to actively learn and optimize the complex trade-offs between identity fidelity, motion naturalness, and temporal coherence. At the heart of our EvoID is a dynamic, dual-path reward mechanism, which acts as an intrinsic critic by adaptively combining objective metric indicators and MLLM-based holistic quality assessment. This allows the model to "evolve" its generation strategy, focusing on different aspects of quality at different stages of training. To ensure stable and coherent evolution, we anchor the exploring Student model with a frozen Teacher, preserving robust world priors while allowing for creative refinement when generating videos. Extensive experiments demonstrate the superiority of our proposal, and EvoID achieves the total score of 0.687 on the Human-Domain of OpenS2V-Eval dataset, surpassing 0.658 of the open-source VACE and 0.653 of the commercial Hailuo. Moreover, EvoID also obtains a new record of 0.718 on our newly minted MLLM-based metric, prioritizing human perception and more comprehensively reflecting video quality.