Jun 2026
ERA: Entropy-Guided Visual Token Pruning with Rectified Attention for Efficient MLLMs
ERA establishes logit-preserving visual token pruning as a principled framework for efficient MLLMs, unifying theoretical foundation, algorithmic design, and practical deployment.
Yuhao Wang, Mu Qiao, Haiwen Diao et al.
· arXiv.org · 0 citations