Video-R4: Reinforcing Text-Rich Video Reasoning with Visual Rumination
Year: 2026
Type: conference-paper
Abstract: Understanding text-rich videos requires reading small, transient textual cues that often demand repeated inspection. Yet most video QA models rely on single-pass perception over fixed frames, leading to hallucinations and failures on finegrained evidence. Inspired by how humans pause, zoom, and re-read critical regions, we introduce Video-R4 (Reinforcing Text-Rich Video Reasoning with Visual Rumination), a video reasoning agent that performs visual rumination: iteratively selecting frames, zooming into informative regions, re-encoding retrieved pixels, and updating its reasoning state. We construct two datasets with executable rumination trajectories: Video-R4-CoT-17k for supervised practice and Video-R4-RL-30k for reinforcement learning. We propose a multi-stage rumination learning framework that progressively finetunes a 7B LMM to learn atomic and mixing visual operations via SFT and GRPO-based RL. Video-R4-7B achieves state-of-the-art results on M4-ViteVQA and further generalizes to multi-page document$Q A$, slides$Q A$, and generic video QA, demonstrating that iterative rumination is an effective paradigm for pixel-grounded multimodal reasoning.
Institutions: University of Rochester, IBM (United States)
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W7220929017