Cover Image

AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhong, Yiwu, Liu, Zhuoming, Li, Yin, Wang, Liwei
Format:	Preprint
Published:	2024
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Rethinking Chain-of-Thought Reasoning for Videos
by: Zhong, Yiwu, et al.
Published: (2025)

Enhancing Temporal Modeling of Video LLMs via Time Gating
by: Hu, Zi-Yuan, et al.
Published: (2024)

PixelPrune: Pixel-Level Adaptive Visual Token Reduction via Predictive Coding
by: Wang, Nan, et al.
Published: (2026)

APLe: Token-Wise Adaptive for Multi-Modal Prompt Learning
by: Cao, Guiming, et al.
Published: (2024)

SpecVLM: Enhancing Speculative Decoding of Video LLMs via Verifier-Guided Token Pruning
by: Ji, Yicheng, et al.
Published: (2025)

LLMs can Compress LLMs: Adaptive Pruning by Agents
by: Kodathala, Sai Varun, et al.
Published: (2026)

PAVE: Patching and Adapting Video Large Language Models
by: Liu, Zhuoming, et al.
Published: (2025)

Beyond Embeddings: The Promise of Visual Table in Visual Reasoning
by: Zhong, Yiwu, et al.
Published: (2024)

Evolutionary Negative Module Pruning for Better LoRA Merging
by: Cao, Anda, et al.
Published: (2026)

LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models
by: Shang, Yuzhang, et al.
Published: (2024)

OccamToken: Efficient VLM Inference with Training-Free and Budget-Adaptive Token Pruning
by: Li, Geng, et al.
Published: (2026)

POP: Prefill-Only Pruning for Efficient Large Model Inference
by: He, Junhui, et al.
Published: (2026)

CoViPAL: Layer-wise Contextualized Visual Token Pruning for Large Vision-Language Models
by: Tang, Zicong, et al.
Published: (2025)

Nüwa: Mending the Spatial Integrity Torn by VLM Token Pruning
by: Huang, Yihong, et al.
Published: (2026)

Towards Learning a Generalist Model for Embodied Navigation
by: Zheng, Duo, et al.
Published: (2023)

MM-R5: MultiModal Reasoning-Enhanced ReRanker via Reinforcement Learning for Document Retrieval
by: Xu, Mingjun, et al.
Published: (2025)

Calibrating Uncertainty Quantification of Multi-Modal LLMs using Grounding
by: Padhi, Trilok, et al.
Published: (2025)

From Pixels to Tokens: Byte-Pair Encoding on Quantized Visual Modalities
by: Zhang, Wanpeng, et al.
Published: (2024)

Index-Preserving Lightweight Token Pruning for Efficient Document Understanding in Vision-Language Models
by: Son, Jaemin, et al.
Published: (2025)

Sparser Block-Sparse Attention via Token Permutation
by: Wang, Xinghao, et al.
Published: (2025)

Less is More: A Simple yet Effective Token Reduction Method for Efficient Multi-modal LLMs
by: Song, Dingjie, et al.
Published: (2024)

RainFusion: Adaptive Video Generation Acceleration via Multi-Dimensional Visual Redundancy
by: Chen, Aiyue, et al.
Published: (2025)

FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token Merging
by: Fan, Ziyang, et al.
Published: (2026)

EVTP-IVS: Effective Visual Token Pruning For Unifying Instruction Visual Segmentation In Multi-Modal Large Language Models
by: Zhu, Wenhui, et al.
Published: (2025)

Multi-Granular Spatio-Temporal Token Merging for Training-Free Acceleration of Video LLMs
by: Hyun, Jeongseok, et al.
Published: (2025)

TokenTrim: Inference-Time Token Pruning for Autoregressive Long Video Generation
by: Shaulov, Ariel, et al.
Published: (2026)

DOSE: Data Selection for Multi-Modal LLMs via Off-the-Shelf Models
by: Wu, Biao, et al.
Published: (2026)

Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic Decoding
by: Xiao, Zhongyu, et al.
Published: (2026)

Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities
by: Yang, Enneng, et al.
Published: (2024)

Core Knowledge Deficits in Multi-Modal Language Models
by: Li, Yijiang, et al.
Published: (2024)

VisCodex: Unified Multimodal Code Generation via Merging Vision and Coding Models
by: Jiang, Lingjie, et al.
Published: (2025)

Bridging the Semantic-Action Gap in Visual Token Pruning for Efficient VLA Inference
by: Liu, Ziyan, et al.
Published: (2025)

Hybrid-Level Instruction Injection for Video Token Compression in Multi-modal Large Language Models
by: Liu, Zhihang, et al.
Published: (2025)

MMICL: Empowering Vision-language Model with Multi-Modal In-Context Learning
by: Zhao, Haozhe, et al.
Published: (2023)

Revisiting Multi-Modal LLM Evaluation
by: Lu, Jian, et al.
Published: (2024)

MANTA: Cross-Modal Semantic Alignment and Information-Theoretic Optimization for Long-form Multimodal Understanding
by: Zhong, Ziqi, et al.
Published: (2025)

An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models
by: Chen, Liang, et al.
Published: (2024)

A Thousand Words or An Image: Studying the Influence of Persona Modality in Multimodal LLMs
by: Broomfield, Julius, et al.
Published: (2025)

VoCoT: Unleashing Visually Grounded Multi-Step Reasoning in Large Multi-Modal Models
by: Li, Zejun, et al.
Published: (2024)

LLMs Can Evolve Continually on Modality for X-Modal Reasoning
by: Yu, Jiazuo, et al.
Published: (2024)