Many video reasoning tasks require tracking motion, temporal order, and evolving visual states across frames. Existing methods built on large vision-language models (LVLMs) often address this challenge by externalizing reasoning through textual chain-of-thought (CoT), keyframe selection, repeated frame reinsertion, or external tool use. While effective, such pipelines increase inference-time latency and engineering complexity, and they force temporal-visual evidence to be serialized into text or repeatedly re-encoded from frames. Inspired by the intuition that visual reasoning can occur implicitly before verbalization, we propose STORMS (Spatial-Temporal reasOning via inteRnalized Modeling), a two-stage framework that teaches LVLMs to reason through bounded continuous latent trajectories instead of explicit textual CoT. In Stage I, STORMS aligns latent tokens with thought-video representations derived from generated videos, grounding the latent states in dynamic visual evidence. In Stage II, the model is further trained with answer-only supervision, encouraging the reasoning process to be internalized without step-by-step annotations. Generated thought videos are used only during training; at inference, STORMS performs a bounded latent rollout without regenerating videos, reinserting frames, or invoking external visual tools. Experiments on VideoMME, MVBench, TempCompass, and MMVU show that STORMS improves video reasoning accuracy while substantially reducing inference overhead compared with tool or video-generation-based reasoning pipelines.

翻译：许多视频推理任务要求跨帧追踪运动、时间顺序及不断演化的视觉状态。现有方法基于大型视觉语言模型（LVLMs），通常通过外部化推理方式来应对这一挑战，例如采用文本思维链（CoT）、关键帧选择、重复帧插入或外部工具调用。尽管有效，此类流程增加了推理延迟和工程复杂性，并迫使时序视觉证据被序列化为文本或反复从帧中重新编码。受视觉推理可在语言化之前隐式完成的直觉启发，我们提出STORMS（通过内化建模实现时空推理），这是一个两阶段框架，教LVLM通过有界的连续潜在轨迹进行推理，而非显式的文本CoT。在第一阶段，STORMS将潜在令牌与从生成视频中推导的思维-视频表征对齐，将潜在状态锚定于动态视觉证据中。在第二阶段，模型进一步接受仅需答案的监督训练，促使推理过程无需逐步标注即可内化。生成思维视频仅用于训练；推理时，STORMS执行有界潜在展开，无需重新生成视频、插入帧或调用外部视觉工具。在VideoMME、MVBench、TempCompass和MMVU上的实验表明，与基于工具或视频生成的推理流程相比，STORMS能提升视频推理准确性，同时显著降低推理开销。