Multimodal large language models (MLLMs) enhance their perceptual capabilities by integrating visual and textual information. However, processing the massive number of visual tokens incurs a significant computational cost. Existing analysis of the MLLM attention mechanisms remains shallow, leading to coarse-grain token pruning strategies that fail to effectively balance speed and accuracy. In this paper, we conduct a comprehensive investigation of MLLM attention mechanisms with LLaVA. We find that numerous visual tokens and partial attention computations are redundant during the decoding process. Based on this insight, we propose Spatial-Temporal Visual Token Trimming ($\textbf{ST}^{3}$), a framework designed to accelerate MLLM inference without retraining. $\textbf{ST}^{3}$ consists of two primary components: 1) Progressive Visual Token Pruning (\textbf{PVTP}), which eliminates inattentive visual tokens across layers, and 2) Visual Token Annealing (\textbf{VTA}), which dynamically reduces the number of visual tokens in each layer as the generated tokens grow. Together, these techniques deliver around $\mathbf{2\times}$ faster inference with only about $\mathbf{30\%}$ KV cache memory compared to the original LLaVA, while maintaining consistent performance across various datasets. Crucially, $\textbf{ST}^{3}$ can be seamlessly integrated into existing pre-trained MLLMs, providing a plug-and-play solution for efficient inference.
Accelerating Multimodal Large Language Models with Prior-Corrected Token Reduction
School of Software Technology, Zhejiang University（浙江大学软件学院）；Zhejiang Key Laboratory of Digital-Intelligence Service Technology（浙江省数字化服务技术重点实验室）；Hong Kong Polytechnic University（香港理工大学）；The University of Southern Queensland（南昆士兰大学）
Accelerating Multimodal Large Language Models via Dynamic Visual-Token Exit and the Empirical Findings
Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University（中国教育部多媒体可信感知与高效计算重点实验室，厦门大学）；Institute of Artificial Intelligence, Xiamen University（厦门大学人工智能研究院）
MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer
School of Information Science and Technology, Fudan University（复旦大学信息科学与技术学院）；Academy for Engineering and Technology, Fudan University（复旦大学工程技术研究院）；Tsinghua-Berkeley Shenzhen Institute, Tsinghua University（清华大学深圳国际研究生院（清华-伯克利深圳学院））
Sparsity Meets Similarity: Leveraging Long-Tail Distribution for Dynamic Optimized Token Representation in Multimodal Large Language Models
School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Institution of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）
iLLaVA: An Image is Worth Fewer Than 1/3 Input Tokens in Large Multimodal Models
School of Computer Science and Technology, Tianjin University（天津大学计算机科学与技术学院）；School of Computer Science and Technology, Tiangong University（天津工大学计算机科学与技术学院）；Key Research Center for Surface Monitoring and Analysis of Relics, State Administration of Cultural Heritage（文物表面监测与分析关键研究中心，国家文物局）
ADEM-VL: Adaptive and Embedded Fusion for Efficient Vision-Language Tuning
School of Information and Electronics, Beijing Institute of Technology（北京理工大学信息与电子学院）；School of Cyber Science and Technology, Sun Yat-sen University（中山大学网络空间安全学院）；School of Computer Science, Wuhan University（武汉大学计算机学院）；School of Computer Science and Engineering, Nanyang Technological University（南洋理工大学计算机科学与工程学院）
VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction
State Key Laboratory for Novel Software Technology, Nanjing University（南京大学新型软件技术国家重点实验室）；School of Intelligence Science and Technology, Nanjing University（南京大学智能科学与技术学院）；Tencent Youtu Lab（腾讯优图实验室）；XMU（厦门大学）；CASIA（中国科学院自动化研究所）
EventFlash: Towards Efficient MLLMs for Event-Based Vision
Xidian University（西安电子科技大学）；Tsinghua University（清华大学）；Beijing Institute of Technology（北京理工大学）；Guangdong Laboratory of Artificial Intelligence and Digital Economy(SZ)（广东人工智能与数字经济实验室（深圳））
HiDrop: Hierarchical Vision Token Reduction in MLLMs via Late Injection, Concave Pyramid Pruning, and Early Exit
Institute of Digital Twin, Eastern Institute of Technology（数字孪生研究所，东部技术研究所）；Ningbo Key Laboratory of Spatial Intelligence and Digital Derivative（宁波空间智能与数字衍生关键实验室）；University of Science and Technology of China（中国科学技术大学）；Shanghai Jiao Tong University（上海交通大学）；Munich Center for Machine Learning, LMU Munich（慕尼黑大学机器学习中心，慕尼黑大学）
We introduce Emu3.5, a large-scale multimodal world model that natively predicts the next state across vision and language. Emu3.5 is pre-trained end-to-end with a unified next-token prediction objective on a corpus of vision-language interleaved data containing over 10 trillion tokens, primarily derived from sequential frames and transcripts of internet videos. The model naturally accepts interleaved vision-language inputs and generates interleaved vision-language outputs. Emu3.5 is further post-trained with large-scale reinforcement learning to enhance multimodal reasoning and generation. To…
Spectral Evolution-Guided Token Pruning in Multimodal Large Language Models
School of Software Technology, Zhejiang University（浙江大学软件学院）；Zhejiang Key Laboratory of Digital-Intelligence Service Technology（浙江省数字化服务技术重点实验室）；The University of Queensland（昆士兰大学）；Singapore Management University（新加坡管理大学）；The University of Southern Queensland（南昆士兰大学）
EntropyPrune: Matrix Entropy Guided Visual Token Pruning for Multimodal Large Language Models
School of Computer Science and Technology, Tongji University（同济大学计算机科学与技术学院）；University of California, Santa Cruz（加州大学圣克ruz分校）；East China Normal University（华东师范大学）；Shanghai Eye Disease Prevention and Treatment Center（上海眼病预防与治疗中心）
Diffusion-based Cumulative Adversarial Purification for Vision Language Models
KTH Royal Institute of Technology（皇家理工学院）；Swiss Federal Institute of Technology Lausanne（洛桑联邦理工学院）；University of California, Los Angeles（加州大学洛杉矶分校）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；CISPA Helmholtz Center for Information Security（信息安全赫尔姆霍兹中心）；RISE Research Institutes of Sweden（瑞典RISE研究机构）；Halmstad University（哈马碧大学）
Fine-Grained Post-Training Quantization for Large Vision Language Models with Quantization-Aware Integrated Gradients
State Key Laboratory of Multimodal Artificial Intelligence Systems, CASIA（多模态人工智能系统国家重点实验室，中国科学院自动化所）；School of Artificial Intelligence, UCAS（人工智能学院，中国科学院大学）；Beijing National Research Center for Information Science and Technology（北京信息科学研究中心）；Institute of Artificial Intelligence, USTB（信息科学技术大学人工智能学院）；School of Artificial Intelligence, Beihang University（北京航空航天大学人工智能学院）；Zhongguancun Academy（中关村学院）
MpoxVLM: A Vision-Language Model for Diagnosing Skin Lesions from Mpox Virus Infection
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of Virginia（弗吉尼亚大学）；Université Quisqueya（基斯克亚大学）；NYU Grossman School of Medicine（纽约大学格罗斯曼医学院）
ScImage: How Good Are Multimodal Large Language Models at Scientific Text-to-Image Generation?
University of Twente（特文特大学）；University of Technology Nuremberg (UTN)（纽伦堡工业大学（UTN））；University of Sheffield（谢菲尔德大学）；Harbin Institute of Technology（哈尔滨工业大学）；University of Mannheim（曼海姆大学）；Technische Universität Dresden（德累斯顿工业大学）
Semantic Equitable Clustering: A Simple and Effective Strategy for Clustering Vision Tokens
MAIS & NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China（自动化研究所，中国科学院，北京）；School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China（中国科学院大学人工智能学院，北京）
Text4Seg++: Advancing Image Segmentation via Generative Language Modeling
College of Computing and Data Science, Nanyang Technological University（computing and Data Science学院，南洋理工大学）；School of Electrical and Electronic Engineering, Nanyang Technological University（Electrical and Electronic Engineering学院，南洋理工大学）；School of Artificial Intelligence, Wuhan University（Artificial Intelligence学院，武汉大学）；ByteDance（字节跳动）
Intern Lumina U2: A Multi-Codebook Diffusion Large Language Model for Omni-Visual Understanding and Image Generation
Shanghai AI Laboratory（上海人工智能实验室）；Shanghai Jiao Tong University（上海交通大学）；Huawei Technologies（华为技术有限公司）；Dalian University of Technology（大连理工大学）；Tsinghua University（清华大学）
