作者位次 #4/8AAAI 2026Poster· Artificial Intelligence
TOP-RL: Task-Optimized Progressive Token Pruning with Reinforcement Learning for Vision Language Models
Hengyi Wang · Weiying Xie · Hui Jiang · Yaotao Wei · Kai Jiang · Mingxiang Cao · Chenhe Hao · Leyuan Fang