Reasoning with Memory: A Temporal Granularity-Adaptive Framework for Training-Free Long Video Understanding
Published in European Conference on Computer Vision (ECCV), 2026
ReMem adapts keyframe selection to a question’s temporal granularity for training-free long-video understanding.
Recommended citation: Linghao Meng, Qiankun Li, Junyuan Mao, Pujin Liao, Zhicheng He, Enbo Zhang, Kun Wang, Yang Liu, Huazhu Fu, Yueming Jin. (2026). "Reasoning with Memory: A Temporal Granularity-Adaptive Framework for Training-Free Long Video Understanding." European Conference on Computer Vision, Pages 588-606.
Download Paper
