[ICLR 2026] AgilePruner: An Empirical Study of Attention and Diversity for Adaptive Visual Token Pruning in Large Vision-Language Models
efficiency attention vlm efficient-inference multimodal hallucination multi-modality token-pruning vision-language-model llava large-vision-language-models qwen lvlm visual-token-pruning iclr2026 agilepruner
-
Updated
Feb 16, 2026