[
    {
        "id": "osp-24019",
        "type": "article-journal",
        "title": "Token-Level Video Reinforcement Learning",
        "author": [
            {
                "family": "Wang",
                "given": "Yifan"
            },
            {
                "family": "Qian",
                "given": "Gordon Guocheng"
            },
            {
                "family": "Li",
                "given": "Yanyu"
            },
            {
                "family": "Kag",
                "given": "Anil"
            },
            {
                "family": "Fu",
                "given": "Yun"
            }
        ],
        "URL": "https://omanscience.com/ar/articles/token-level-video-reinforcement-learning",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Reinforcement learning (RL) for video generation usually assigns one scalar reward to an entire sampled video. Yet a video is not uniformly flawed: some visual tokens may already satisfy the prompt, whereas others require correction. A scalar reward cannot localize errors, causing optimization to perturb satisfactory tokens while under-targeting the tokens that actually need to change. We introduce Token-Level Video Reinforcement Learning, TVRL, a framework that derives token-level credit from the reward being optimized. Our key insight is that the answer likelihood of a frozen vision-language model provides both signals: its outputs contribute to the video-level reward, while magnitudes of its video-input gradients reveal which generated video tokens most affect that score. We instantiate TVRL in Group Relative Policy Optimization by averaging prompt-derived question rewards into one group-relative advantage and using detached, question-conditioned token-credit maps to reweight dense denoising-transition log-probabilities inside the clipped policy ratio. On VBench-2.0, TVRL achieves an Overall score of 57.69, outperforming the base model by 3.60 points. TVRL also improves matched GRPO baselines across three SDE samplers (SAGE, Flow, and Dance) by 2.68--3.15 points and across four reward models (VideoAlign, VideoScore2, UnifiedReward2, and Qwen3.5-9B) by 1.33--3.15 points."
    }
]