# MASSV: Multimodal Adaptation and Self-Data Distillation for Speculative Decoding of Vision-Language Models

arXiv: https://arxiv.org/abs/2505.10526 · HTML: https://mlsanity.com/paper/2505.10526/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 26 of its 38 references that are in the map.
- First posted on arXiv 15 May 2025.

## 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 / [Fast autoregressive visual generation](https://mlsanity.com/inference/fast-autoregressive-visual-generation/)

## Builds on (most cited first, 26 of 26)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023)
- [StarCoder: may the source be with you!](https://mlsanity.com/paper/2305.06161/) (2023)
- [EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty](https://mlsanity.com/paper/2401.15077/) (2024)
- [EAGLE-2: Faster Inference of Language Models with Dynamic Draft Trees](https://mlsanity.com/paper/2406.16858/) (2024)
- [LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding](https://mlsanity.com/paper/2404.16710/) (2024)
- [REST: Retrieval-Based Speculative Decoding](https://mlsanity.com/paper/2311.08252/) (2023)
- [Self-Distillation Bridges Distribution Gap in Language Model Fine-Tuning](https://mlsanity.com/paper/2402.13669/) (2024)
- [SWIFT: On-the-Fly Self-Speculative Decoding for LLM Inference Acceleration](https://mlsanity.com/paper/2410.06916/) (2024)
- [OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure](https://mlsanity.com/paper/2406.17276/) (2024)
- [Visual Program Distillation: Distilling Tools and Programmatic Reasoning into Vision-Language Models](https://mlsanity.com/paper/2312.03052/) (2023)
- [Inference with Reference: Lossless Acceleration of Large Language Models](https://mlsanity.com/paper/2304.04487/) (2023)
- [On Speculative Decoding for Multimodal Large Language Models](https://mlsanity.com/paper/2404.08856/) (2024)
- [Lossless Acceleration of Large Language Model via Adaptive N-gram Parallel Decoding](https://mlsanity.com/paper/2404.08698/) (2024)
- [Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies](https://mlsanity.com/paper/2502.05202/) (2025)
- [The N-Grammys: Accelerating Autoregressive Inference with Learning-Free Batched Speculation](https://mlsanity.com/paper/2411.03786/) (2024)
- [Self-Data Distillation for Recovering Quality in Pruned Large Language Models](https://mlsanity.com/paper/2410.09982/) (2024)
- [On Teacher Hacking in Language Model Distillation](https://mlsanity.com/paper/2502.02671/) (2025)
- [SD2: Self-Distilled Sparse Drafters](https://mlsanity.com/paper/2504.08838/) (2025)

## Built on by (most cited first, 5 of 5)

- [AsymSpec: Context-Asymmetric Speculative Decoding for Agentic LLMs](https://mlsanity.com/paper/2608.26004/) (2026)
- [EvoSpec: Evolving Speculative Decoding via Real-Time Vocabulary and Parameter Adaptation](https://mlsanity.com/paper/2605.27390/) (2026)
- [Is Multimodal Speculative Decoding Ready for Diffusion-Based Parallel Drafting? A Survey and Empirical Diagnosis](https://mlsanity.com/paper/2608.20743/) (2026)
- [HAWK: Rethinking Multimodal Drafting for Speculative Decoding](https://mlsanity.com/paper/2610.00623/) (2026)
- [SP2ec: Adaptive Self-Speculative Decoding for Vision-Language Models](https://mlsanity.com/paper/2610.18735/) (2026)

## Abstract (arXiv)

Speculative decoding significantly accelerates language model inference by enabling a lightweight draft model to propose multiple tokens that a larger target model verifies simultaneously. However, applying this technique to vision-language models (VLMs) presents two fundamental challenges: small language models that could serve as efficient drafters lack the architectural components to process visual inputs, and their token predictions fail to match those of VLM target models that consider visual context. We introduce Multimodal Adaptation and Self-Data Distillation for Speculative Decoding of Vision-Language Models (MASSV), which transforms existing small language models into effective multimodal drafters through a two-phase approach. MASSV first connects the target VLM's vision encoder to the draft model via a lightweight trainable projector, then applies self-distilled visual instruction tuning using responses generated by the target VLM to align token predictions. Comprehensive 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. MASSV provides a scalable, architecture-compatible method for accelerating both current and future VLMs.