# Adaptive Two-Stage Visual Token Pruning for Efficient Inference in Video-Language Models

arXiv: https://arxiv.org/abs/2608.03112 · HTML: https://mlsanity.com/paper/2608.03112/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 27 of its 32 references that are in the map.
- First posted on arXiv 4 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)
- 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, 27 of 27)

- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs](https://mlsanity.com/paper/2406.07476/) (2024)
- [NExT-QA:Next Phase of Question-Answering to Explaining Temporal Actions](https://mlsanity.com/paper/2105.08276/) (2021)
- [An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models](https://mlsanity.com/paper/2403.06764/) (2024)
- [LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models](https://mlsanity.com/paper/2311.17043/) (2023)
- [LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models](https://mlsanity.com/paper/2403.15388/) (2024)
- [LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding](https://mlsanity.com/paper/2410.17434/) (2024)
- [Perception Test: A Diagnostic Benchmark for Multimodal Video Models](https://mlsanity.com/paper/2305.13786/) (2023)
- [DivPrune: Diversity-based Visual Token Pruning for Large Multimodal Models](https://mlsanity.com/paper/2503.02175/) (2025)
- [TinyLLaVA: A Framework of Small-scale Large Multimodal Models](https://mlsanity.com/paper/2402.14289/) (2024)
- [Efficient Multimodal Learning from Data-centric Perspective](https://mlsanity.com/paper/2402.11530/) (2024)
- [ATP-LLaVA: Adaptive Token Pruning for Large Vision Language Models](https://mlsanity.com/paper/2412.00447/) (2024)
- [Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters](https://mlsanity.com/paper/2406.12335/) (2024)
- [TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model](https://mlsanity.com/paper/2503.18278/) (2025)
- [Video-XL-Pro: Reconstructive Token Compression for Extremely Long Video Understanding](https://mlsanity.com/paper/2503.18478/) (2025)
- [LLaVA-MoD: Making LLaVA Tiny via MoE Knowledge Distillation](https://mlsanity.com/paper/2408.15881/) (2024)
- [LLaVA-KD: A Framework of Distilling Multimodal Large Language Models](https://mlsanity.com/paper/2410.16236/) (2024)
- [Filter, Correlate, Compress: Training-Free Token Reduction for MLLM Acceleration](https://mlsanity.com/paper/2411.17686/) (2024)
- [AlphaPruning: Using Heavy-Tailed Self Regularization Theory for Improved Layer-wise Pruning of Large Language Models](https://mlsanity.com/paper/2410.10912/) (2024)
- [Dynamic-VLM: Simple Dynamic Visual Token Compression for VideoLLM](https://mlsanity.com/paper/2412.09530/) (2024)
- [LLaVA-Scissor: Token Compression with Semantic Connected Components for Video LLMs](https://mlsanity.com/paper/2506.21862/) (2025)
- [Imp: Highly Capable Large Multimodal Models for Mobile Devices](https://mlsanity.com/paper/2405.12107/) (2024)

## Abstract (arXiv)

Vision-language models excel at image and video understanding but suffer from high inference latency due to the need to process thousands of tokens per image, limiting their deployment on resource-constrained edge devices and in real-time surveillance applications. This challenge is further amplified in video processing, where multiple frames must be analyzed simultaneously. Existing token reduction techniques are largely developed for single-image inputs and therefore fail to account for the temporal and inter-frame redundancies present in video sequences. In addition, these methods generally rely on a fixed, uniform pruning ratio applied across all inputs, which is suboptimal because the degree of redundancy can vary significantly between different videos, necessitating content-dependent pruning levels to preserve critical information. To address these limitations, we propose a two-stage adaptive token pruning strategy specifically designed for video processing. In the first stage, we prune out the redundant frames, and in the second stage, token-level pruning is applied within the retained frames. Crucially, the pruning ratio in the second stage is determined adaptively based on the content of each video. This is achieved by analyzing the correlation structure of token embeddings to quantify redundancy, which is used to determine the ratio. Importantly, our method is entirely post-hoc and requires no additional training or fine-tuning, while achieving strong empirical gains; notably, it improves accuracy by +7% on a video captioning benchmark at 10% token retention, while reducing computation TFLOPs by 95%.