Reasoning via Video: The First Evaluation of Video Models' Reasoning Abilities through Maze-Solving Tasks

TL;DR

This paper introduces VR-Bench, a new benchmark for evaluating video models' reasoning abilities through maze-solving tasks, demonstrating their spatial reasoning strengths and the benefits of diverse sampling during inference.

Contribution

The paper presents VR-Bench, the first comprehensive benchmark for reasoning via video, and empirically evaluates video models' spatial reasoning capabilities using maze-solving tasks.

Findings

Video models outperform VLMs in spatial reasoning tasks.

Diverse sampling during inference improves reasoning reliability by 10-20%.

Video models generalize well across different maze types and complexities.

Abstract

Video Models have achieved remarkable success in high-fidelity video generation with coherent motion dynamics. Analogous to the development from text generation to text-based reasoning in language modeling, the development of video models motivates us to ask: Can video models reason via video generation? Compared with the discrete text corpus, video grounds reasoning in explicit spatial layouts and temporal continuity, which serves as an ideal substrate for spatial reasoning. In this work, we explore the reasoning via video paradigm and introduce VR-Bench -- a comprehensive benchmark designed to systematically evaluate video models' reasoning capabilities. Grounded in maze-solving tasks that inherently require spatial planning and multi-step reasoning, VR-Bench contains 7,920 procedurally generated videos across five maze types and diverse visual styles. Our empirical analysis demonstrates that SFT can efficiently elicit the reasoning ability of video model. Video models exhibit stronger spatial perception during reasoning, outperforming leading VLMs and generalizing well across diverse scenarios, tasks, and levels of complexity. We further discover a test-time scaling effect, where diverse sampling during inference improves reasoning reliability by 10--20%. These findings highlight the unique potential and scalability of reasoning via video for spatial reasoning tasks.