Worse Than Random: Understanding Visual Token Information in VLLMs," Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2026. [Bolya+ 2023] D. Bolya et al., "Token Merging: Your ViT but Faster," Proc. Int'l Conf. Learning Representations (ICLR), 2023. [Li+ 2023] J. Li et al., "BLIP-2: Bootstrapping Language-Image Pre-Training with Frozen Image Encoders and Large Language Models," Proc. 40th Int'l Conf. Machine Learning (ICML), 2023, pp. 19730–19742. [Chen+ 2024] L. Chen et al., "An Image Is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models," Proc. European Conf. Computer Vision (ECCV), 2024, pp. 19–35. [Liu+ 2024] H. Liu et al., "Improved Baselines with Visual Instruction Tuning," Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2024, pp. 26296–26306. [Alvar+ 2025] S.R. Alvar et al., "DivPrune: Diversity-Based Visual Token Pruning for Large Multimodal Models," Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2025, pp. 9392–9401. [Bai+ 2025] S. Bai et al., "Qwen2.5-VL Technical Report," arXiv:2502.13923, 2025. [Lin+ 2025] Z. Lin et al., "Boosting Multimodal Large Language Models with Visual Tokens Withdrawal for Rapid Inference," Proc. AAAI Conf. Artificial Intelligence (AAAI), 2025, pp. 5334–5342. [Wen+ 2025] Z. Wen et al., "Stop Looking for Important Tokens in Multimodal Language Models: Duplication Matters More," Proc. Conf. Empirical Methods in Natural Language Processing (EMNLP), 2025.