Image de couverture de livre

Improving Visual Token Reduction via Rectifying Distortions for Efficient Multimodal LLM Inference

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Cho, Hyeonwoo, Baek, DongHyeon, Kim, Yewon, Ham, Bumsub
Format:	Preprint
Publié:	2026
Sujets:	Computer Vision and Pattern Recognition
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

FYI: Flip Your Images for Dataset Distillation
par: Son, Byunggwan, et autres
Publié: (2024)

Efficient Few-Shot Neural Architecture Search by Counting the Number of Nonlinear Functions
par: Oh, Youngmin, et autres
Publié: (2024)

GrowTAS: Progressive Expansion from Small to Large Subnets for Efficient ViT Architecture Search
par: Lee, Hyunju, et autres
Publié: (2025)

AZ-NAS: Assembling Zero-Cost Proxies for Network Architecture Search
par: Lee, Junghyup, et autres
Publié: (2024)

Maximizing the Position Embedding for Vision Transformers with Global Average Pooling
par: Lee, Wonjun, et autres
Publié: (2025)

TAS-LoRA: Transformer Architecture Search with Mixture-of-LoRA Experts
par: Jeon, Jeimin, et autres
Publié: (2026)

Scheduling Weight Transitions for Quantization-Aware Training
par: Lee, Junghyup, et autres
Publié: (2024)

Toward INT4 Fixed-Point Training via Exploring Quantization Error for Gradients
par: Kim, Dohyung, et autres
Publié: (2024)

Subnet-Aware Dynamic Supernet Training for Neural Architecture Search
par: Jeon, Jeimin, et autres
Publié: (2025)

Slot-MLLM: Object-Centric Visual Tokenization for Multimodal LLM
par: Chi, Donghwan, et autres
Publié: (2025)

Instance-Aware Group Quantization for Vision Transformers
par: Moon, Jaehyeon, et autres
Publié: (2024)

Jailbreaking on Text-to-Video Models via Scene Splitting Strategy
par: Lee, Wonjun, et autres
Publié: (2025)

Exploring Hierarchical Consistency and Unbiased Objectness for Open-Vocabulary Object Detection
par: Lee, Sanghoon, et autres
Publié: (2026)

3DPillars: Pillar-based two-stage 3D object detection
par: Noh, Jongyoun, et autres
Publié: (2025)

Cerberus: Attribute-based person re-identification using semantic IDs
par: Eom, Chanho, et autres
Publié: (2024)

TokenPacker: Efficient Visual Projector for Multimodal LLM
par: Li, Wentong, et autres
Publié: (2024)

DeVI: Physics-based Dexterous Human-Object Interaction via Synthetic Video Imitation
par: Kim, Hyeonwoo, et autres
Publié: (2026)

Disentangled Representations for Short-Term and Long-Term Person Re-Identification
par: Eom, Chanho, et autres
Publié: (2024)

Relational Feature Caching for Accelerating Diffusion Transformers
par: Son, Byunggwan, et autres
Publié: (2026)

Joint-Embedding Predictive Architecture for Self-Supervised Learning of Mask Classification Architecture
par: Kim, Dong-Hee, et autres
Publié: (2024)

CNG-SFDA:Clean-and-Noisy Region Guided Online-Offline Source-Free Domain Adaptation
par: Cho, Hyeonwoo, et autres
Publié: (2024)

TRIO: Token Reduction via Inference-Objective Guidance for Efficient Vision-Language Models
par: Zhang, Haokui, et autres
Publié: (2026)

AccuQuant: Simulating Multiple Denoising Steps for Quantizing Diffusion Models
par: Lee, Seunghoon, et autres
Publié: (2025)

DisCoRD: Discrete Tokens to Continuous Motion via Rectified Flow Decoding
par: Cho, Jungbin, et autres
Publié: (2024)

VScan: Rethinking Visual Token Reduction for Efficient Large Vision-Language Models
par: Zhang, Ce, et autres
Publié: (2025)

VisionSelector: End-to-End Learnable Visual Token Compression for Efficient Multimodal LLMs
par: Zhu, Jiaying, et autres
Publié: (2025)

Frequency-Aware Token Reduction for Efficient Vision Transformer
par: Lee, Dong-Jae, et autres
Publié: (2025)

Efficient Rectified Flow for Image Fusion
par: Wang, Zirui, et autres
Publié: (2025)

Understanding and Rectifying Safety Perception Distortion in VLMs
par: Zou, Xiaohan, et autres
Publié: (2025)

ELITE: Enhanced Language-Image Toxicity Evaluation for Safety
par: Lee, Wonjun, et autres
Publié: (2025)

Robustness-Reinforced Knowledge Distillation with Correlation Distance and Network Pruning
par: Kim, Seonghak, et autres
Publié: (2023)

MVFormer: Diversifying Feature Normalization and Token Mixing for Efficient Vision Transformers
par: Bae, Jongseong, et autres
Publié: (2024)

SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference
par: Zhang, Yuan, et autres
Publié: (2024)

EchoingPixels: Cross-Modal Adaptive Token Reduction for Efficient Audio-Visual LLMs
par: Gong, Chao, et autres
Publié: (2025)

v1: Learning to Point Visual Tokens for Multimodal Grounded Reasoning
par: Chung, Jiwan, et autres
Publié: (2025)

Embedding-Free Transformer with Inference Spatial Reduction for Efficient Semantic Segmentation
par: Yu, Hyunwoo, et autres
Publié: (2024)

G$^2$TR: Generation-Guided Visual Token Reduction for Separate-Encoder Unified Multimodal Models
par: Li, Junxian, et autres
Publié: (2026)

GlocalCLIP: Object-agnostic Global-Local Prompt Learning for Zero-shot Anomaly Detection
par: Ham, Jiyul, et autres
Publié: (2024)

Two Birds, One Projection: Harmonizing Safety and Utility in LVLMs via Inference-time Feature Projection
par: Han, Yewon, et autres
Publié: (2026)

Learning 3D Object Spatial Relationships from Pre-trained 2D Diffusion Models
par: Baik, Sangwon, et autres
Publié: (2025)