Speculative Decoding Reimagined for Multimodal Large Language Models

• URL: http://arxiv.org/abs/2505.14260v1
• Date: Tue, 20 May 2025 12:12:17 GMT
• Title: Speculative Decoding Reimagined for Multimodal Large Language Models
• Authors: Luxi Lin, Zhihang Lin, Zhanpeng Zeng, Rongrong Ji,
• Abstract summary: This paper introduces Multimodal Speculative Decoding (MSD) to accelerate Multimodal Large Language Models (MLLMs) inference.<n>Experiments show that MSD boosts inference speed by up to $2.29times$ for LLaVA-1.5-7B and up to $2.46times$ for LLaVA-1.5-13B on multimodal benchmarks.
• Score: 48.115777709178595
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper introduces Multimodal Speculative Decoding (MSD) to accelerate Multimodal Large Language Models (MLLMs) inference. Speculative decoding has been shown to accelerate Large Language Models (LLMs) without sacrificing accuracy. However, current speculative decoding methods for MLLMs fail to achieve the same speedup as they do for LLMs. To address this, we reimagine speculative decoding specifically for MLLMs. Our analysis of MLLM characteristics reveals two key design principles for MSD: (1) Text and visual tokens have fundamentally different characteristics and need to be processed separately during drafting. (2) Both language modeling ability and visual perception capability are crucial for the draft model. For the first principle, MSD decouples text and visual tokens in the draft model, allowing each to be handled based on its own characteristics. For the second principle, MSD uses a two-stage training strategy: In stage one, the draft model is trained on text-only instruction-tuning datasets to improve its language modeling ability. In stage two, MSD gradually introduces multimodal data to enhance the visual perception capability of the draft model. Experiments show that MSD boosts inference speed by up to $2.29\times$ for LLaVA-1.5-7B and up to $2.46\times$ for LLaVA-1.5-13B on multimodal benchmarks, demonstrating its effectiveness. Our code is available at https://github.com/Lyn-Lucy/MSD.

Related papers

• MASSV: Multimodal Adaptation and Self-Data Distillation for Speculative Decoding of Vision-Language Models [0.09895793818721334]
  We introduce Multimodal Adaptation and Self-Data Distillation for Speculative Decoding of Vision-Language Models (MASSV)<n>MASSV transforms existing small language models into effective multimodal drafters through a two-phase approach.<n>Experiments across the Qwen2.5-VL and Gemma3 model families demonstrate that MASSV increases accepted length by up to 30% and delivers end-to-end inference speedups of up to 1.46x on visually-grounded tasks.
  arXiv  Detail & Related papers  (2025-05-15T17:37:00Z)
• TAMP: Token-Adaptive Layerwise Pruning in Multimodal Large Language Models [23.916205754112774]
  Multimodal Large Language Models (MLLMs) have shown remarkable versatility in understanding diverse multimodal data and tasks.<n>We propose TAMP, a simple yet effective pruning framework tailored for MLLMs.<n>We validate our method on two state-of-the-art MLLMs: LLaVA-NeXT, designed for vision-language tasks, and VideoLLaMA2, capable of processing audio, visual, and language modalities.
  arXiv  Detail & Related papers  (2025-04-14T05:44:38Z)
• Liquid: Language Models are Scalable and Unified Multi-modal Generators [112.71734051183726]
  Liquid is an auto-regressive generation paradigm that seamlessly integrates visual comprehension and generation.<n>Unlike previous multimodal large language model (MLLM), Liquid achieves this integration using a single large language model.<n>For the first time, Liquid uncovers a scaling law that performance drop unavoidably brought by the unified training of visual and language tasks.
  arXiv  Detail & Related papers  (2024-12-05T16:48:16Z)
• Looking Beyond Text: Reducing Language bias in Large Vision-Language Models via Multimodal Dual-Attention and Soft-Image Guidance [67.26434607115392]
  Large vision-language models (LVLMs) have achieved impressive results in various vision-language tasks. LVLMs suffer from hallucinations caused by language bias, leading to diminished focus on images and ineffective visual comprehension. We propose LACING to address the language bias of LVLMs with muLtimodal duAl-attention meChanIsm (MDA) aNd soft-image Guidance (IFG)
  arXiv  Detail & Related papers  (2024-11-21T16:33:30Z)
• LLaVA-KD: A Framework of Distilling Multimodal Large Language Models [72.68665884790002]
  We propose a novel framework to transfer knowledge from l-MLLMs to s-MLLMs.<n>We introduce Multimodal Distillation (MDist) to transfer teacher model's robust representations across both visual and linguistic modalities.<n>We also propose a three-stage training scheme to fully exploit the potential of the proposed distillation strategy.
  arXiv  Detail & Related papers  (2024-10-21T17:41:28Z)
• NVLM: Open Frontier-Class Multimodal LLMs [64.00053046838225]
  We introduce NVLM 1.0, a family of frontier-class multimodal large language models (LLMs) that achieve state-of-the-art results on vision-language tasks. We propose a novel architecture that enhances both training efficiency and multimodal reasoning capabilities. We develop production-grade multimodality for the NVLM-1.0 models, enabling them to excel in vision-language tasks.
  arXiv  Detail & Related papers  (2024-09-17T17:59:06Z)
• NoteLLM-2: Multimodal Large Representation Models for Recommendation [71.87790090964734]
  Large Language Models (LLMs) have demonstrated exceptional proficiency in text understanding and embedding tasks.<n>Their potential in multimodal representation, particularly for item-to-item (I2I) recommendations, remains underexplored.<n>We propose an end-to-end fine-tuning method that customizes the integration of any existing LLMs and vision encoders for efficient multimodal representation.
  arXiv  Detail & Related papers  (2024-05-27T03:24:01Z)
• On Speculative Decoding for Multimodal Large Language Models [11.245862832561176]
  Inference with Multimodal Large Language Models (MLLMs) is slow due to their large-language-model backbone. We show that a language-only model can serve as a good draft model for speculative decoding with LLaVA 7B.
  arXiv  Detail & Related papers  (2024-04-13T00:02:36Z)

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.