Dynamic Token Reduction during Generation for Vision Language Models

• URL: http://arxiv.org/abs/2501.14204v1
• Date: Fri, 24 Jan 2025 03:20:37 GMT
• Title: Dynamic Token Reduction during Generation for Vision Language Models
• Authors: Xiaoyu Liang, Chaofeng Guan, Jiaying Lu, Huiyao Chen, Huan Wang, Haoji Hu,
• Abstract summary: We introduce a dynamic pruning strategy tailored for Vision-Language Models (VLMs)<n>Our approach enables flexible adjustment of pruning rates based on the attention distribution.<n>Our experimental results demonstrate that our method not only reduces computational demands but also maintains the quality of responses.
• Score: 11.376359442815986
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-Language Models (VLMs) have achieved notable success in multimodal tasks but face practical limitations due to the quadratic complexity of decoder attention mechanisms and autoregressive generation. Existing methods like FASTV and VTW have achieved notable results in reducing redundant visual tokens, but these approaches focus on pruning tokens in a single forward pass without systematically analyzing the redundancy of visual tokens throughout the entire generation process. In this paper, we introduce a dynamic pruning strategy tailored for VLMs, namedDynamic Rate (DyRate), which progressively adjusts the compression rate during generation. Our analysis of the distribution of attention reveals that the importance of visual tokens decreases throughout the generation process, inspiring us to adopt a more aggressive compression rate. By integrating a lightweight predictor based on attention distribution, our approach enables flexible adjustment of pruning rates based on the attention distribution. Our experimental results demonstrate that our method not only reduces computational demands but also maintains the quality of responses.

Related papers

• DyMU: Dynamic Merging and Virtual Unmerging for Efficient VLMs [124.52164183968145]
  We present DyMU, an efficient, training-free framework that reduces the computational burden of vision-language models (VLMs) Our approach comprises two key components. First, Dynamic Token Merging (DToMe) reduces the number of visual token embeddings by merging similar tokens based on image complexity. Second, Virtual Token Unmerging (VTU) simulates the expected token sequence for large language models (LLMs) by efficiently reconstructing the attention dynamics of a full sequence.
  arXiv  Detail & Related papers  (2025-04-23T18:38:18Z)
• Saliency-driven Dynamic Token Pruning for Large Language Models [32.903622070917194]
  Saliency-driven Dynamic Token Pruning (SDTP) A lightweight saliency-driven prediction module is designed to estimate the importance score of each token with its hidden state. A ranking-based optimization strategy is proposed to minimize the ranking divergence of the saliency score and the predicted importance score.
  arXiv  Detail & Related papers  (2025-04-06T15:15:07Z)
• Skip-Vision: Efficient and Scalable Acceleration of Vision-Language Models via Adaptive Token Skipping [13.846838416902575]
  A key bottleneck stems from the proliferation of visual tokens required for fine-grained image understanding. We propose Skip-Vision, a unified framework addressing both training and inference inefficiencies in vision-language models. Experimental results demonstrate that Skip-Vision reduces training time by up to 35%, inference FLOPs by 75%, and latency by 45%.
  arXiv  Detail & Related papers  (2025-03-26T04:16:48Z)
• Bridging Continuous and Discrete Tokens for Autoregressive Visual Generation [63.89280381800457]
  We propose TokenBridge, which maintains the strong representation capacity of continuous tokens while preserving the modeling simplicity of discrete tokens. We introduce a dimension-wise quantization strategy that independently discretizes each feature dimension, paired with a lightweight autoregressive prediction mechanism. Our approach achieves reconstruction and generation quality on par with continuous methods while using standard categorical prediction.
  arXiv  Detail & Related papers  (2025-03-20T17:59:59Z)
• MINT: Mitigating Hallucinations in Large Vision-Language Models via Token Reduction [6.416957959150438]
  Hallucinations hinder the application of Large Vision-Language Models (LVLMs) in domains that require high reliability. We propose MINT, a training-free decoding strategy, MItigating hallucinations via tokeN reducTion. Our approach achieves a 4% improvement in mitigating hallucinations caused by distracted perception compared to original models.
  arXiv  Detail & Related papers  (2025-02-02T08:34:57Z)
• AdaFV: Accelerating VLMs with Self-Adaptive Cross-Modality Attention Mixture [7.9213473377478865]
  Some approaches to reduce the visual tokens according to the self-attention of VLMs, which are biased, result in inaccurate responses.<n>We propose a self-adaptive cross-modality attention mixture mechanism that dynamically leverages the effectiveness of visual saliency and text-to-image similarity.<n>The proposed approach achieves state-of-the-art training-free VLM acceleration performance, especially when the reduction rate is sufficiently large.
  arXiv  Detail & Related papers  (2025-01-16T13:34:33Z)
• Video Token Sparsification for Efficient Multimodal LLMs in Autonomous Driving [9.900979396513687]
  Multimodal large language models (MLLMs) have demonstrated remarkable potential for enhancing scene understanding in autonomous driving systems. One major limitation arises from the large number of visual tokens required to capture fine-grained and long-context visual information. We propose Video Token Sparsification (VTS) to significantly reduce the total number of visual tokens while preserving the most salient information.
  arXiv  Detail & Related papers  (2024-09-16T05:31:01Z)
• 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)
• Efficient Diffusion Transformer with Step-wise Dynamic Attention Mediators [83.48423407316713]
  We present a novel diffusion transformer framework incorporating an additional set of mediator tokens to engage with queries and keys separately. Our model initiates the denoising process with a precise, non-ambiguous stage and gradually transitions to a phase enriched with detail. Our method achieves a state-of-the-art FID score of 2.01 when integrated with the recent work SiT.
  arXiv  Detail & Related papers  (2024-08-11T07:01:39Z)
• Efficient Large Multi-modal Models via Visual Context Compression [23.966237939194514]
  We present the study on the analysis of redundancy concerning visual tokens and efficient training within large language models. Our initial experiments show that eliminating up to 70% of visual tokens at the testing stage by simply average pooling only leads to a minimal 3% reduction in visual question answering accuracy. We introduce Visual Context on the GQA benchmark, which reduces the number of visual tokens to enhance training and inference efficiency without sacrificing performance.
  arXiv  Detail & Related papers  (2024-06-28T17:57:14Z)
• DynaSeg: A Deep Dynamic Fusion Method for Unsupervised Image Segmentation Incorporating Feature Similarity and Spatial Continuity [0.5755004576310334]
  We introduce DynaSeg, an innovative unsupervised image segmentation approach. Unlike traditional methods, DynaSeg employs a dynamic weighting scheme that adapts flexibly to image characteristics. DynaSeg prevents undersegmentation failures where the number of predicted clusters might converge to one.
  arXiv  Detail & Related papers  (2024-05-09T00:30:45Z)
• Accelerating Vision Transformers Based on Heterogeneous Attention Patterns [89.86293867174324]
  Vision Transformers (ViTs) have attracted a lot of attention in the field of computer vision. We propose an integrated compression pipeline based on observed heterogeneous attention patterns across layers. Experimentally, the integrated compression pipeline of DGSSA and GLAD can accelerate up to 121% run-time throughput.
  arXiv  Detail & Related papers  (2023-10-11T17:09:19Z)
• Uncovering the Hidden Cost of Model Compression [43.62624133952414]
  Visual Prompting has emerged as a pivotal method for transfer learning in computer vision. Model compression detrimentally impacts the performance of visual prompting-based transfer. However, negative effects on calibration are not present when models are compressed via quantization.
  arXiv  Detail & Related papers  (2023-08-29T01:47:49Z)
• SmartTrim: Adaptive Tokens and Attention Pruning for Efficient Vision-Language Models [35.5601603013045]
  We propose SmartTrim, an adaptive acceleration framework for Vision-Language Models (VLMs) We integrate lightweight modules into the original backbone to identify and prune redundant token representations and attention heads within each layer. We devise a self-distillation strategy to enhance the consistency between the predictions of the pruned model and its fully-capacity counterpart.
  arXiv  Detail & Related papers  (2023-05-24T11:18:00Z)
• Perceiver-VL: Efficient Vision-and-Language Modeling with Iterative Latent Attention [100.81495948184649]
  We present Perceiver-VL, a vision-and-language framework that efficiently handles high-dimensional multimodal inputs such as long videos and text. Our framework scales with linear complexity, in contrast to the quadratic complexity of self-attention used in many state-of-the-art transformer-based models.
  arXiv  Detail & Related papers  (2022-11-21T18:22:39Z)
• Untangling tradeoffs between recurrence and self-attention in neural networks [81.30894993852813]
  We present a formal analysis of how self-attention affects gradient propagation in recurrent networks. We prove that it mitigates the problem of vanishing gradients when trying to capture long-term dependencies. We propose a relevancy screening mechanism that allows for a scalable use of sparse self-attention with recurrence.
  arXiv  Detail & Related papers  (2020-06-16T19:24:25Z)

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.