1 paper
Junzhe Chen, Siyuan Meng, Yuxi Chen +3
Video large language models (Video-LLMs) have made strong progress in general video understanding, but their ability to maintain temporal object consistency remains underexplored.…