ReMoT: Reinforcement Learning with Motion Contrast Triplets
Wan, Cong, Guo, Zeyu, Li, Jiangyang, Dong, Songlin, Bai, Yifan, Peng, Lin, Ma, Zhiheng, Gong, Yihong
原文摘要
We present ReMoT, a unified training paradigm to systematically address the fundamental shortcomings of VLMs in spatio-temporal consistency--a critical failure point in navigation, robotics, and autonomous driving. ReMoT integrates two core components: (i) A rule-based automatic framework that generates ReMoT-16K, a large-scale (16.5K triplets) motion-contrast dataset derived from video meta-annotations, surpassing costly manual or model-based generation. (ii) Group Relative Policy Optimization, which we empirically validate, yields optimal performance and data efficiency for learning this contrastive reasoning, far exceeding standard Supervised Fine-Tuning. We also construct the first benchmark for fine-grained motion contrast triplets to measure a VLM's discrimination of subtle motion attributes (e.g., opposing directions). The resulting model achieves SOTA performance on our new benchmark and multiple standard VLM benchmarks, culminating in a remarkable 25.1 performance leap on spatio-temporal reasoning tasks.
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表