PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction

• URL: http://arxiv.org/abs/2410.17247v2
• Date: Thu, 27 Feb 2025 11:16:33 GMT
• Title: PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction
• Authors: Long Xing, Qidong Huang, Xiaoyi Dong, Jiajie Lu, Pan Zhang, Yuhang Zang, Yuhang Cao, Conghui He, Jiaqi Wang, Feng Wu, Dahua Lin,
• Abstract summary: In large vision-language models (LVLMs), images serve as inputs that carry a wealth of information.<n>Previous approaches have attempted to reduce the number of image tokens either before or within the early layers of LVLMs.<n>We propose PyramidDrop, a visual redundancy reduction strategy for LVLMs to boost their efficiency in both training and inference.
• Score: 94.11897755903519
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In large vision-language models (LVLMs), images serve as inputs that carry a wealth of information. As the idiom "A picture is worth a thousand words" implies, representing a single image in current LVLMs can require hundreds or even thousands of tokens. This results in significant computational costs, which grow quadratically as input image resolution increases, thereby severely impacting the efficiency of both training and inference. Previous approaches have attempted to reduce the number of image tokens either before or within the early layers of LVLMs. However, these strategies inevitably result in the loss of crucial image information, ultimately diminishing model performance. To address this challenge, we conduct an empirical study revealing that all visual tokens are necessary for LVLMs in the shallow layers, and token redundancy progressively increases in the deeper layers of the model. To this end, we propose PyramidDrop, a visual redundancy reduction strategy for LVLMs to boost their efficiency in both training and inference with neglectable performance loss. Specifically, we partition the LVLM into several stages and drop part of the image tokens at the end of each stage with a pre-defined ratio, creating pyramid-like visual tokens across model layers. The dropping is based on a lightweight similarity calculation with a negligible time overhead. Extensive experiments demonstrate that PyramidDrop can achieve a 40% training time and 55% inference FLOPs acceleration of LLaVA-NeXT with comparable performance. Besides, the PyramidDrop could also serve as a plug-and-play strategy for inference acceleration without training, with better performance and lower inference cost than counterparts. Code is available at https://github.com/Cooperx521/PyramidDrop.

Related papers

• LVLM_CSP: Accelerating Large Vision Language Models via Clustering, Scattering, and Pruning for Reasoning Segmentation [10.614327633823462]
  Large Vision Language Models (LVLMs) have been widely adopted to guide vision foundation models in performing reasoning segmentation tasks. In this work, we propose LVLM_CSP, a novel training free visual token pruning method specifically designed for LVLM based reasoning segmentation tasks.
  arXiv  Detail & Related papers  (2025-04-15T04:27:15Z)
• Accelerating Multimodal Large Language Models by Searching Optimal Vision Token Reduction [62.8375542401319]
  Multimodal Large Language Models (MLLMs) encode the input image(s) as vision tokens and feed them into the language backbone. The number of vision tokens increases quadratically as the image resolutions, leading to huge computational costs. We propose a greedy search algorithm (G-Search) to find the least number of vision tokens to keep at each layer from the shallow to the deep.
  arXiv  Detail & Related papers  (2024-11-30T18:54:32Z)
• FoPru: Focal Pruning for Efficient Large Vision-Language Models [11.36025001578531]
  We propose Focal Pruning (FoPru), a training-free method that prunes visual tokens based on the attention-based token significance derived from the vision encoder. Our method can prune a large number of redundant tokens while maintaining high accuracy, leading to significant improvements in inference efficiency.
  arXiv  Detail & Related papers  (2024-11-21T14:22:38Z)
• VideoLLM-MoD: Efficient Video-Language Streaming with Mixture-of-Depths Vision Computation [66.00245701441547]
  We introduce a novel approach to reduce vision compute by leveraging redundant vision tokens "skipping layers" rather than decreasing the number of vision tokens. Our method, VideoLLM-MoD, is inspired by mixture-of-depths LLMs and addresses the challenge of numerous vision tokens in long-term or streaming video.
  arXiv  Detail & Related papers  (2024-08-29T17:21:58Z)
• Paying More Attention to Image: A Training-Free Method for Alleviating Hallucination in LVLMs [14.381188702947949]
  Large Vision-Language Models (LVLMs) primarily align image features of vision encoder with Large Language Models (LLMs) to leverage their superior text generation capabilities. This imbalance in LVLMs may result in the instances of hallucinatory. We introduce a training-free algorithm to find an equilibrium point between image comprehension and language inference.
  arXiv  Detail & Related papers  (2024-07-31T17:46:57Z)
• Data Attribution for Text-to-Image Models by Unlearning Synthesized Images [71.23012718682634]
  The goal of data attribution for text-to-image models is to identify the training images that most influence the generation of a new image. We propose a new approach that efficiently identifies highly-influential images.
  arXiv  Detail & Related papers  (2024-06-13T17:59:44Z)
• An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models [65.37846460916042]
  We find out that the attention computation over visual tokens is of extreme inefficiency in the deep layers of popular LVLMs. We introduce FastV, a versatile plug-and-play method designed to optimize computational efficiency.
  arXiv  Detail & Related papers  (2024-03-11T14:35:32Z)
• ELIP: Efficient Language-Image Pre-training with Fewer Vision Tokens [75.09406436851445]
  We propose a vision token pruning and merging method ELIP, to remove less influential tokens based on the supervision of language outputs. Our experiments demonstrate that with the removal of 30$%$ vision tokens across 12 ViT layers, ELIP maintains significantly comparable performance.
  arXiv  Detail & Related papers  (2023-09-28T05:31:07Z)
• Memory Efficient Meta-Learning with Large Images [62.70515410249566]
  Meta learning approaches to few-shot classification are computationally efficient at test time requiring just a few optimization steps or single forward pass to learn a new task. This limitation arises because a task's entire support set, which can contain up to 1000 images, must be processed before an optimization step can be taken. We propose LITE, a general and memory efficient episodic training scheme that enables meta-training on large tasks composed of large images on a single GPU.
  arXiv  Detail & Related papers  (2021-07-02T14:37:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.