PACE: A Unified Condense-and-Extract Paradigm for Fast VLM Inference

arXiv:2608.27206 · cs.CV, cs.AI · Submitted 2026-08-27 · Read on arXiv

cs.CV, cs.AI

Submitted: 2026-08-27

Updated: 2026-09-21

Code: https://github.com/jjL357/PACE

Terminology

• Pruning — 29× in this paper · explained in 4 episodes
• Benchmark — 25× in this paper · explained in 2 episodes
• Redundancy — 14× in this paper · explained in 2 episodes
• Accuracy — 13× in this paper · explained in 1 episode
• Ablation — 5× in this paper · explained in 1 episode
• Large Language Model — 5× in this paper
• Robustness — 5× in this paper · explained in 5 episodes
• Vision-language models — 5× in this paper · explained in 2 episodes
• Adaptive Pixel Compressor — 4× in this paper
• Aggregation — 4× in this paper · explained in 1 episode
• Alignment — 4× in this paper · explained in 1 episode
• Transformer — 4× in this paper · explained in 2 episodes
• Quantization — 3× in this paper · explained in 12 episodes
• Core — 2× in this paper · explained in 2 episodes
• Generalization — 2× in this paper · explained in 13 episodes
• Hallucination — 2× in this paper · explained in 3 episodes
• Instruction Tuning — 2× in this paper · explained in 2 episodes
• Vision-Language Models (VLMs) — 2× in this paper · explained in 6 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Amortization — 1× in this paper · explained in 1 episode
• Attention Mechanism — 1× in this paper · explained in 3 episodes
• Computational overhead — 1× in this paper · explained in 2 episodes
• Cosine Similarity — 1× in this paper · explained in 3 episodes
• Decode — 1× in this paper · explained in 2 episodes
• Dynamic Dual-Attention Extractor — 1× in this paper
• Edge Devices — 1× in this paper · explained in 2 episodes
• Embeddings — 1× in this paper · explained in 2 episodes
• F1 Score — 1× in this paper · explained in 3 episodes
• Feature Extraction — 1× in this paper · explained in 2 episodes
• Few-Shot Learning — 1× in this paper · explained in 4 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Large Language Model (LLM) — 1× in this paper · explained in 3 episodes
• Multimodal Reasoning — 1× in this paper · explained in 5 episodes
• Predictive Coding — 1× in this paper · explained in 2 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Structural Reasoning — 1× in this paper · explained in 2 episodes

Sources

• Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
• GPT-4 Technical Report
• DocPrune:Efficient Document Question Answering via Background, Question, and Comprehension-aware Token Pruning
• Accelerating Vision Transformers with Adaptive Patch Sizes
• ViCO: A Training Strategy towards Semantic Aware Dynamic High-Resolution
• Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond
• Qwen3-VL Technical Report
• Qwen2.5-VL Technical Report
• MMTok: Multimodal Coverage Maximization for Efficient Inference of VLMs
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Prune Redundancy, Preserve Essence: Vision Token Compression in VLMs via Synergistic Importance-Diversity
• Seed1.5-VL Technical Report
• Are We Using the Right Benchmark: An Evaluation Framework for Visual Token Compression Methods
• AdaptVision: Efficient Vision-Language Models via Adaptive Visual Acquisition
• LearnPruner: Rethinking Attention-based Token Pruning in Vision Language Models
• Gemini: A Family of Highly Capable Multimodal Models
• HyperVL: An Efficient and Dynamic Multimodal Large Language Model for Edge Devices
• TwinQuant: Learnable Subspace Decomposition for 4-Bit LLM Quantization
• PixelPrune: Pixel-Level Adaptive Visual Token Reduction via Predictive Coding
• SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models