ReMoRa: Multimodal Large Language Model based on Refined Motion Representation for Long-Video Understanding

Daichi Yashima, Shuhei Kurita, Yusuke Oda, Komei Sugiura; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026, pp. 31845-31855

While multimodal large language models (MLLMs) have shown remarkable success across a wide range of tasks, long-form video understanding remains a significant challenge.In this study, we focus on video understanding by MLLMs.This task is challenging because processing a full stream of RGB frames is computationally intractable and highly redundant, as self-attention have quadratic complexity with sequence length.In this paper, we propose ReMoRa, a video MLLM that processes videos by operating directly on their compressed representations.A sparse set of RGB keyframes is retained for appearance, while temporal dynamics are encoded as a motion representation, removing the need for sequential RGB frames.These motion representations act as a compact proxy for optical flow, capturing temporal dynamics without full frame decoding.To refine the noise and low fidelity of block-based motions, we introduce a module to denoise and generate a fine-grained motion representation.Furthermore, our model compresses these features in a way that scales linearly with sequence length.We demonstrate the effectiveness of ReMoRa through extensive experiments across a comprehensive suite of long-video understanding benchmarks.ReMoRa outperformed baseline methods on multiple challenging benchmarks, including LongVideoBench, NExT-QA, and MLVU. Our project page is available at https://remora-v1rcm.kinsta.page/.