# CROP: Contextual Region-Oriented Visual Token Pruning

arXiv: https://arxiv.org/abs/2505.21233 · HTML: https://mlsanity.com/paper/2505.21233/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Visual token pruning in multimodal models, one of 12 directions in Efficient LLM inference and serving.
- Builds on 24 of its 46 references that are in the map.
- First posted on arXiv 27 May 2025.

## Placed in

- Efficient LLM inference and serving / [Visual token pruning in multimodal models](https://mlsanity.com/inference/visual-token-pruning-in-multimodal-models/)

## Builds on (most cited first, 24 of 24)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI](https://mlsanity.com/paper/2311.16502/) (2023)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Towards VQA Models That Can Read](https://mlsanity.com/paper/1904.08920/) (2019)
- [V*: Guided Visual Search as a Core Mechanism in Multimodal LLMs](https://mlsanity.com/paper/2312.14135/) (2023)
- [TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding](https://mlsanity.com/paper/2312.02051/) (2023)
- [SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference](https://mlsanity.com/paper/2410.04417/) (2024)
- [PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction](https://mlsanity.com/paper/2410.17247/) (2024)
- [MLVU: Benchmarking Multi-task Long Video Understanding](https://mlsanity.com/paper/2406.04264/) (2024)
- [Visual CoT: Advancing Multi-Modal Language Models with a Comprehensive Dataset and Benchmark for Chain-of-Thought Reasoning](https://mlsanity.com/paper/2403.16999/) (2024)
- [MLLMs Know Where to Look: Training-free Perception of Small Visual Details with Multimodal LLMs](https://mlsanity.com/paper/2502.17422/) (2025)
- [LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token](https://mlsanity.com/paper/2501.03895/) (2025)
- [Dynamic-LLaVA: Efficient Multimodal Large Language Models via Dynamic Vision-language Context Sparsification](https://mlsanity.com/paper/2412.00876/) (2024)
- [LLaVA-Interactive: An All-in-One Demo for Image Chat, Segmentation, Generation and Editing](https://mlsanity.com/paper/2311.00571/) (2023)
- [Filter, Correlate, Compress: Training-Free Token Reduction for MLLM Acceleration](https://mlsanity.com/paper/2411.17686/) (2024)
- [TextCoT: Zoom In for Enhanced Multimodal Text-Rich Image Understanding](https://mlsanity.com/paper/2404.09797/) (2024)
- [KTAE: A Model-Free Algorithm to Key-Tokens Advantage Estimation in Mathematical Reasoning](https://mlsanity.com/paper/2505.16826/) (2025)
- [Learning Musical Representations for Music Performance Question Answering](https://mlsanity.com/paper/2502.06710/) (2025)
- [LADM: Long-context Training Data Selection with Attention-based Dependency Measurement for LLMs](https://mlsanity.com/paper/2503.02502/) (2025)
- [TG-LLaVA: Text Guided LLaVA via Learnable Latent Embeddings](https://mlsanity.com/paper/2409.09564/) (2024)

## Built on by (most cited first, 14 of 14)

- [Look Again, Think Slowly: Enhancing Visual Reflection in Vision-Language Models](https://mlsanity.com/paper/2509.12132/) (2025)
- [Teaching Vision-Language Models to Ask: Resolving Ambiguity in Visual Questions](https://mlsanity.com/paper/2507.13773/) (2025)
- [ACE-RL: Adaptive Constraint-Enhanced Reward for Long-form Generation Reinforcement Learning](https://mlsanity.com/paper/2509.04903/) (2025)
- [Look Where It Matters: High-Resolution Crops Retrieval for Efficient VLMs](https://mlsanity.com/paper/2603.16932/) (2026)
- [CoVSpec: Efficient Device-Edge Co-Inference for Vision-Language Models via Speculative Decoding](https://mlsanity.com/paper/2605.02218/) (2026)
- [RUTA: Principled Visual Token Allocation via Rate-Utility Optimization](https://mlsanity.com/paper/2608.04132/) (2026)
- [Discovering Semantic Subdimensions through Disentangled Conceptual Representations](https://mlsanity.com/paper/2508.21436/) (2025)
- [When RAG Hurts: Diagnosing and Mitigating Attention Distraction in Retrieval-Augmented LVLMs](https://mlsanity.com/paper/2602.00344/) (2026)
- [Baseline-Relative Counterfactual Refinement for Bit-Aware Visual Token Communication](https://mlsanity.com/paper/2608.16192/) (2026)
- [ProtoVQA: An Adaptable Prototypical Framework for Explainable Fine-Grained Visual Question Answering](https://mlsanity.com/paper/2509.16680/) (2025)
- [VideoRouter: Query-Adaptive Dual Routing for Efficient Long-Video Understanding](https://mlsanity.com/paper/2605.05848/) (2026)
- [Beyond Accuracy: Auditing Spatial Provenance in Visual Token Pruning for OCR-Critical MLLM Inference](https://mlsanity.com/paper/2608.00077/) (2026)
- [Don't Send What You Don't Need: Question-Guided Token Pruning as a Privacy Defense for Vision-Language Models](https://mlsanity.com/paper/2609.15671/) (2026)
- [BrainFocus: EEG-Guided ROI Selection for Efficient Vision-Language Models](https://mlsanity.com/paper/2609.17443/) (2026)

## Abstract (arXiv)

Current VLM-based VQA methods often process entire images, leading to excessive visual tokens that include redundant information irrelevant to the posed question. This abundance of unnecessary image details creates numerous visual tokens, drastically increasing memory and computational requirements in VLMs. To address this, we propose Contextual Region-Oriented Visual Token Pruning (CROP), a novel framework to compress visual tokens through a two-step process: Localization and Pruning. Specifically, CROP first employs an efficient model to identify the contextual region relevant to the input query. Subsequently, two distinct strategies are introduced for pruning: (1) Pre-LLM Compression (PLC), which adaptively compresses different image regions with varying ratios, and (2) Inner-LLM Pruning (ILP), a training-free method that prunes tokens within early LLM layers guided by the identified contextual region. Extensive experiments on a wide range of VQA tasks demonstrate that CROP significantly outperforms existing visual token pruning methods and achieves state-of-the-art performance.