Overview

Visual token pruning studies how to remove or merge image and video patches before or inside vision transformers and vision-language models, reducing computation and memory for tasks such as visual question answering, video understanding, document reading, and robot control. Methods use attention or text-query relevance, feature diversity and spatial coverage, and signals such as entropy, spectral structure, or geometry; learned selectors and adaptive budgets choose different tokens or compression rates for different inputs. Pruning can happen before visual encoding, across language-model layers, or during video processing, and is evaluated on both task benchmarks and practical measures such as prefill latency, FLOPs, and memory. The central tension is that a token that seems redundant early may become important for later reasoning, while selection overhead and hardware execution can prevent fewer tokens from yielding faster end-to-end inference.

Latest papers

CardsList

298 papers23 papers in the last four weeks, down 48% on the four weeks before. 0.2% of all new papers.

  1. GSTEP: Global Spatio-Temporal Density-Driven Visual Token Pruning for Efficient Video Large Language Models

    Aug 4, 2026Mengjie Zhang, Qihui Zhu, Tao Zhang +10Efficient VLM InferenceVideo-Language Models

  2. Messages, Not Tokens: Grounded Coresets for Faithful VLM Compression

    Aug 3, 2026Long Qian, Jiaqi Wei, Bingke Zhu +2Efficient VLM InferenceCoreset Selection

  3. DiffPrune: differentiable information throttling for token pruning in vision-language models

    Aug 3, 2026Landi He, Mingde Yao, Shawn Young +1Vision-Language ModelsEfficient VLM Inference

  4. ET-Prune: Evidence-Aware Dynamic Budgeting for Visual Token Pruning in Text-Rich MLLMs

    Aug 3, 2026Zizhong Ding, Junxian Li, Kai Liu +4Efficient Multimodal InferenceMultimodal Large Language Models

  5. Think in Sets for Streaming Video Token Compression

    Aug 2, 2026Moxu Duan, Jingwen Fu, Yuwang WangVideo Token CompressionVisual Token Pruning

  6. The Gate, Not the Cache: Gate Provenance Bounds the Closed-Loop Reliability of Training-Free VLA Token Skipping

    Aug 1, 2026Qi Luo, Shuaijun Liu, Hao Zhao +5Efficient VLA Model InferenceVision-Language-Action Models

  7. CodeShrink: Adaptive Visual Compression for Efficient Multimodal Code Understanding

    Jul 31, 2026Wenxin Tang, Jingyu Xiao, Zhenyu Liu +6Efficient Multimodal InferenceMultimodal Large Language Models

  8. ReToken: Improving Long-Context VLMs with Visual Retrieval Token

    Jul 30, 2026Yao Xiao, Reuben Tan, Zhen Zhu +3Vision-Language ModelsVision-Language Retrieval

  9. Capturing Token Tendencies for Training-Free Token Pruning in Multimodal Large Language Models

    Jul 30, 2026Jie Ma, Zhike Qiu, Jie Gao +4Efficient Multimodal InferenceMultimodal Large Language Models

  10. LAST: The Last Query Token Guides Visual Token Pruning for Edge-Cloud Collaborative MLLM Inference

    Jul 30, 2026Feng Yang, Xinrui Ju, Keyang Zhang +6Edge-Cloud Collaborative InferenceEfficient VLM Inference

  11. Beyond Accuracy: Auditing Spatial Provenance in Visual Token Pruning for OCR-Critical MLLM Inference

    Jul 29, 2026Feixiang Liu, Qiang Qiu, Hao Zhang +1VLM EvaluationEfficient VLM Inference

  12. SepPrune:A Separator-based Pruning Framework for Efficient Multimodal Large Language Models

    Jul 28, 2026Yuchen Wang, Qihui Zhu, Yang Liu +2Efficient Multimodal InferenceMultimodal Large Language Models

  13. GOTS: Greedy Orthogonal Token Selection for High-Resolution Vision-Language Models

    Jul 27, 2026Jun Ling, Tao Huang, Junzhuo Liu +2Efficient VLM InferenceLarge Vision-Language Models

  14. PathSelect: Sequential Token Selection for Whole Slide Pathology

    Jul 26, 2026Jingzhi Chen, Landi He, Zehong Chen +2Efficient VLM InferenceMedical VLMs

  15. OmniScope: Modality-Decoupled Token Compression for Omnimodal Large Language Models

    Jul 25, 2026Jinsen Su, Yongdong Luo, Yuexiao Ma +3Efficient Multimodal InferenceAudio Token Compression

  16. Structured Redundancy Modeling for Efficient Visual Token Pruning in High-Resolution MLLMs

    Jul 25, 2026Jouwon Song, Woohyeong Kim, Kyeongbo KongEfficient VLM InferenceLarge Vision-Language Models

  17. LayoutLite: Token-Level Implicit Layout Analysis for Efficient Document OCR

    Jul 24, 2026Xudong Liu, Bicheng Wan, Yulin JinEfficient VLM InferenceDocument Layout Analysis

  18. Out of Sight, Still in Mind: Token Compression for Omni-LLMs

    Jul 23, 2026Suho Yoo, Youngjoon Jang, Hyebin Cho +1Efficient Multimodal InferenceMultimodal Large Language Models

  19. Look Less, Think Faster: Joint Token-Compute Adaptation for Multimodal LLMs

    Jul 22, 2026Pengcheng Wang, Zhiquan Wang, Jayoung Lee +5Efficient Multimodal InferenceEfficient VLM Inference

  20. Visual Token Compression Enhances Robustness of MLLMs

    Jul 21, 2026Shishen Gu, Jiequan Cui, Wenbo Hu +3VLM RobustnessAdversarial Attacks on VLMs

  21. BMFA: Boundary-Minority Free-Energy Adaptive Screening

    Jul 20, 2026Wenyan Xu, Alizer WongSoftmax AttentionVision Transformer

  22. Brain-Aligned Multi-Stream Video Transformers with Sparse Self-Selection

    Jul 20, 2026Amir Hosein Fadaei, Mahyar Maleki, Mohammad-Reza A. DehaqaniTransformerEfficient ViTs

  23. Are All Tokens Necessary for Visual Place Recognition? An Empirical Study of Token Reduction for Efficient Inference

    Jul 17, 2026Tong Jin, Yunpeng Liu, Shuyu Hu +4Visual Place RecognitionEfficient ViTs

  24. Attention-Free and Lightweight Token Reduction for Efficient Vision-Language Models

    Jul 15, 2026Xuanyi Hao, Zuoyuan Zhang, Zhibo Wang +4Vision-Language ModelsEfficient VLM Inference

  25. Spectral Heat Flow for Conservative Token Condensation in Vision-Language Models

    Jul 12, 2026Zhaoyang Li, Yanjun Li, Wangkai Li +2Efficient VLM InferenceVisual Token Pruning

  26. Foveation-Guided Dynamic Token Selection for Robust and Efficient Vision Transformers

    Jul 10, 2026Ibrahim Batuhan Akkaya, Kishaan Jeeveswaran, Bahram Zonooz +1Image Corruption RobustnessEfficient ViTs

  27. TORINO: Token Reduction via Interpretable Concept Overlap in Vision-Language Models

    Jul 6, 2026Riccardo Renzulli, Gabriele Spadaro, Shruthi Gowda +2Efficient VLM InferenceVisual Token Pruning

  28. Seeing Once is Enough? Online Geometry-Aware Token Pruning for 3D Question Answering

    Jul 5, 2026Ruei-Chi Lai, Bolivar Solarte, Chin-Hsuan Wu +2Efficient VLM InferenceMultimodal QA

  29. Combating Textual Noise and Redundancy: Entropy-Aware Dense Visual Token Pruning

    Jul 2, 2026Xuehui Wang, Xuankun Yang, Wei ShenEfficient VLM InferenceMultimodal Token Compression