HiViS: Hiding Visual Tokens from the Drafter for Speculative Decoding in Vision-Language Models

• URL: http://arxiv.org/abs/2509.23928v1
• Date: Sun, 28 Sep 2025 15:05:21 GMT
• Title: HiViS: Hiding Visual Tokens from the Drafter for Speculative Decoding in Vision-Language Models
• Authors: Zhinan Xie, Peisong Wang, Jian Cheng,
• Abstract summary: We propose Hiding Visual Tokens from the Drafter for Speculative Decoding in Vision-Language Models (HiViS)<n>HiViS is an explicit-implicit input decomposition framework that alleviates the inefficiency of Speculative Decoding in Vision-Language Models.<n>Our approach compresses the prefill sequence length of the drafter to only 0.7%-1.3% of the target VLM's input.
• Score: 23.98782884568504
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Speculative decoding is an effective approach for accelerating inference in Large Language models (LLMs), but its adaptation to Vision-Language models (VLMs) remains challenging for additional visual tokens in multimodal inputs. First, owing to the fact that the drafter and the target VLM may derived from different families, the semantic representations of visual tokens in the target VLM are misaligned with those in the drafter, introducing bias into the KV-cache during the prefill stage. Second, the large number of visual tokens substantially slows down the drafter's self-attention during the decoding stage. We propose Hiding Visual Tokens from the Drafter for Speculative Decoding in Vision-Language Models (HiViS), an explicit-implicit input decomposition framework that alleviates the above inefficiency. All visual tokens are removed from the drafter's input, retaining only textual tokens as explicit inputs, while directly reusing the target VLM's corresponding last-layer hidden states as implicit visual information without additional processing. To train the drafter efficiently, we introduces multi-step self-feedback training strategy with dynamic data selection and sequential embedding supervision to simulate reasoning during training. Our approach compresses the prefill sequence length of the drafter to only 0.7%-1.3% of the target VLM's input, while maintaining lossless generation quality. Extensive experiments across diverse models and tasks demonstrate up to 2.65x speedup, confirming the effectiveness of HiViS in accelerating VLM inference.

Related papers

• CoViPAL: Layer-wise Contextualized Visual Token Pruning for Large Vision-Language Models [75.88232735646018]
  Large Vision-Language Models (LVLMs) process multimodal inputs consisting of text tokens and vision tokens extracted from images or videos.<n>Existing methods attempt to prune redundant vision tokens, revealing substantial redundancy in visual representations.<n>We propose CoViPAL, a layer-wise contextualized visual token pruning method that employs a Plug-and-Play Pruning Module (PPM) to predict and remove redundant vision tokens before they are processed by the LVLM.
  arXiv  Detail & Related papers  (2025-08-24T07:47:00Z)
• Rethinking Visual Token Reduction in LVLMs under Cross-modal Misalignment [38.04426918886084]
  Vision-Language Models (LVLMs) encode visual inputs as dense sequences of patch-level tokens to capture fine-grained semantics.<n>Previous efforts have explored visual token reduction either prior to or within the large language models (LLMs)<n>We introduce VisionDrop, a training-free, visual-only pruning framework that selects informative visual tokens based on intra-modal (visual-to-visual) attention.
  arXiv  Detail & Related papers  (2025-06-27T14:55:40Z)
• Revisit What You See: Disclose Language Prior in Vision Tokens for Efficient Guided Decoding of LVLMs [8.97780713904412]
  This paper introduces ReVisiT, a simple yet effective decoding method that references vision tokens to guide the text generation process in Large Vision-Language Models (LVLMs)<n>Our approach leverages the semantic information embedded within vision tokens by projecting them into the text token distribution space, and dynamically selecting the most relevant vision token at each decoding step through constrained divergence minimization. Experiments on three LVLM benchmarks with two recent LVLMs demonstrate that ReVisiT consistently enhances visual grounding with minimal computational overhead.
  arXiv  Detail & Related papers  (2025-06-11T08:46:55Z)
• VScan: Rethinking Visual Token Reduction for Efficient Large Vision-Language Models [57.2662376527586]
  VScan is a two-stage visual token reduction framework.<n>It addresses token redundancy by: (1) integrating complementary global and local scans with token merging during visual encoding, and (2) introducing pruning at intermediate layers of the language model.<n>VScan achieves a 2.91$times$ speedup in prefilling and a 10$times$ reduction in FLOPs, while retaining 95.4% of the original performance.
  arXiv  Detail & Related papers  (2025-05-28T17:59:08Z)
• Semantic-Clipping: Efficient Vision-Language Modeling with Semantic-Guidedd Visual Selection [53.558449071113245]
  Vision-Language Models (VLMs) leverage aligned visual encoders to transform images into visual tokens, allowing them to be processed similarly to text by the backbone large language model (LLM)<n>Recent advancements in vision-language modeling introduce image cropping techniques that feed all encoded sub-images into the model.<n>We propose a lightweight, universal framework that seamlessly integrates with existing VLMs to enhance their ability to process finegrained details.
  arXiv  Detail & Related papers  (2025-03-14T18:33:31Z)
• D-Attn: Decomposed Attention for Large Vision-and-Language Models [29.611769371733672]
  We propose Decomposed Attention (D-Attn), a more flexible attention architecture for large vision-and-language models (LVLMs)<n>D-Attn decomposes the 1-D causal self-attention of LVLMs into visual-to-visual, textual-to-visual, and textual-to-textual attentions.<n>Experiments and analysis validate the effectiveness of D-Attn, demonstrating significant improvements on multiple image benchmarks.
  arXiv  Detail & Related papers  (2025-02-04T00:46:11Z)
• AdaFV: Rethinking of Visual-Language alignment for VLM acceleration [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)
• 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)
• Towards Semantic Equivalence of Tokenization in Multimodal LLM [149.11720372278273]
  Vision tokenization is essential for semantic alignment between vision and language.<n>This paper proposes a novel dynamic Semantic-Equivalent Vision Tokenizer (SeTok)<n>SeTok groups visual features into semantic units via a dynamic clustering algorithm.<n>The resulting vision tokens effectively preserve semantic integrity and capture both low-frequency and high-frequency visual features.
  arXiv  Detail & Related papers  (2024-06-07T17:55:43Z)
• Memory-Space Visual Prompting for Efficient Vision-Language Fine-Tuning [59.13366859237086]
  Current solutions for efficiently constructing large vision-language (VL) models follow a two-step paradigm. We consider visual prompts as additional knowledge that facilitates language models in addressing tasks associated with visual information. We introduce a novel approach, wherein visual prompts are memoryd with the weights of FFN for visual knowledge injection.
  arXiv  Detail & Related papers  (2024-05-09T08:23:20Z)
• Unified Language-Vision Pretraining in LLM with Dynamic Discrete Visual Tokenization [52.935150075484074]
  We introduce a well-designed visual tokenizer to translate the non-linguistic image into a sequence of discrete tokens like a foreign language. The resulting visual tokens encompass high-level semantics worthy of a word and also support dynamic sequence length varying from the image. This unification empowers LaVIT to serve as an impressive generalist interface to understand and generate multi-modal content simultaneously.
  arXiv  Detail & Related papers  (2023-09-09T03:01:38Z)

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.