Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoning

• URL: http://arxiv.org/abs/2601.21037v1
• Date: Wed, 28 Jan 2026 20:57:55 GMT
• Title: Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoning
• Authors: Chengzu Li, Zanyi Wang, Jiaang Li, Yi Xu, Han Zhou, Huanyu Zhang, Ruichuan An, Dengyang Jiang, Zhaochong An, Ivan Vulić, Serge Belongie, Anna Korhonen,
• Abstract summary: We formulate visual reasoning by means of video generation models.<n>We evaluate their capacity in two distinct regimes: Maze Navigation for sequential discrete planning with low visual change and Tangram Puzzle for continuous manipulation with high visual change.
• Score: 38.651924340946785
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision-Language Models have excelled at textual reasoning, but they often struggle with fine-grained spatial understanding and continuous action planning, failing to simulate the dynamics required for complex visual reasoning. In this work, we formulate visual reasoning by means of video generation models, positing that generated frames can act as intermediate reasoning steps between initial states and solutions. We evaluate their capacity in two distinct regimes: Maze Navigation for sequential discrete planning with low visual change and Tangram Puzzle for continuous manipulation with high visual change. Our experiments reveal three critical insights: (1) Robust Zero-Shot Generalization: In both tasks, the model demonstrates strong performance on unseen data distributions without specific finetuning. (2) Visual Context: The model effectively uses visual context as explicit control, such as agent icons and tangram shapes, enabling it to maintain high visual consistency and adapt its planning capability robustly to unseen patterns. (3) Visual Test-Time Scaling: We observe a test-time scaling law in sequential planning; increasing the generated video length (visual inference budget) empowers better zero-shot generalization to spatially and temporally complex paths. These findings suggest that video generation is not merely a media tool, but a scalable, generalizable paradigm for visual reasoning.

Related papers

• STARCaster: Spatio-Temporal AutoRegressive Video Diffusion for Identity- and View-Aware Talking Portraits [44.82339975771063]
  STARCaster is an identity-aware video diffusion model that addresses both speech-driven portrait animation and free-point-view talking portrait.<n>The model learns from longer temporal contexts than those generated at inference, mitigating the overly static animations common in existing autoregressive approaches.
  arXiv  Detail & Related papers  (2025-12-15T11:59:01Z)
• Video-R2: Reinforcing Consistent and Grounded Reasoning in Multimodal Language Models [56.851611990473174]
  Reasoning over dynamic visual content remains a central challenge for large language models.<n>We propose a reinforcement learning approach that enhances both temporal precision and reasoning consistency.<n>The resulting model, Video R2, achieves consistently higher TAC, VAS, and accuracy across multiple benchmarks.
  arXiv  Detail & Related papers  (2025-11-28T18:59:58Z)
• Plan-X: Instruct Video Generation via Semantic Planning [36.020841550221824]
  Plan-X is a framework that explicitly enforces high-level semantic planning to instruct video generation process.<n>Our framework substantially reduces visual hallucinations and enables fine-grained, instruction-aligned video generation consistent with multimodal context.
  arXiv  Detail & Related papers  (2025-11-22T08:59:09Z)
• Show Me: Unifying Instructional Image and Video Generation with Diffusion Models [16.324312147741495]
  We propose a unified framework that enables image manipulation and video prediction.<n>We introduce structure and motion consistency rewards to improve structural fidelity and temporal coherence.<n> Experiments on diverse benchmarks demonstrate that our method outperforms expert models in both instructional image and video generation.
  arXiv  Detail & Related papers  (2025-11-21T23:24:28Z)
• FrameMind: Frame-Interleaved Video Reasoning via Reinforcement Learning [65.42201665046505]
  Current video understanding models rely on fixed frame sampling strategies, processing predetermined visual inputs regardless of the specific reasoning requirements of each question.<n>This static approach limits their ability to adaptively gather visual evidence, leading to suboptimal performance on tasks that require broad temporal coverage or fine-grained spatial detail.<n>We introduce FrameMind, an end-to-end framework trained with reinforcement learning that enables models to dynamically request visual information during reasoning through Frame-Interleaved Chain-of-Thought (FiCOT)<n>Unlike traditional approaches, FrameMind operates in multiple turns where the model alternates between textual reasoning and active visual perception, using tools to extract
  arXiv  Detail & Related papers  (2025-09-28T17:59:43Z)
• MiCo: Multi-image Contrast for Reinforcement Visual Reasoning [72.81576836419373]
  Chain-of-Thought (CoT) reasoning can be used to link visual cues across multiple images.<n>We adapt rule-based reinforcement learning for Vision-Language Models (VLMs)<n>Our method achieves significant improvements on multi-image reasoning benchmarks and shows strong performance on general vision tasks.
  arXiv  Detail & Related papers  (2025-06-27T17:59:27Z)
• Understanding Long Videos via LLM-Powered Entity Relation Graphs [51.13422967711056]
  GraphVideoAgent is a framework that maps and monitors the evolving relationships between visual entities throughout the video sequence.<n>Our approach demonstrates remarkable effectiveness when tested against industry benchmarks.
  arXiv  Detail & Related papers  (2025-01-27T10:57:24Z)
• Video as Conditional Graph Hierarchy for Multi-Granular Question Answering [80.94367625007352]
  We argue that while video is presented in frame sequence, the visual elements are not sequential but rather hierarchical in semantic space. We propose to model video as a conditional graph hierarchy which weaves together visual facts of different granularity in a level-wise manner.
  arXiv  Detail & Related papers  (2021-12-12T10:35:19Z)
• MERLOT: Multimodal Neural Script Knowledge Models [74.05631672657452]
  We introduce MERLOT, a model that learns multimodal script knowledge by watching millions of YouTube videos with transcribed speech. MERLOT exhibits strong out-of-the-box representations of temporal commonsense, and achieves state-of-the-art performance on 12 different video QA datasets. On Visual Commonsense Reasoning, MERLOT answers questions correctly with 80.6% accuracy, outperforming state-of-the-art models of similar size by over 3%.
  arXiv  Detail & Related papers  (2021-06-04T17:57:39Z)
• Local Frequency Domain Transformer Networks for Video Prediction [24.126513851779936]
  Video prediction is of interest not only in anticipating visual changes in the real world but has, above all, emerged as an unsupervised learning rule. This paper proposes a fully differentiable building block that can perform all of those tasks separately while maintaining interpretability.
  arXiv  Detail & Related papers  (2021-05-10T19:48:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.