AASD: Accelerate Inference by Aligning Speculative Decoding in Multimodal Large Language Models
Keywords: Multimodal Large Language Model, Inference Acceleration, Speculative Decoding
Abstract: Multimodal Large Language Models (MLLMs) have achieved notable success in visual instruction tuning, yet their inference is time-consuming due to the auto-regressive decoding of Large Language Model (LLM) backbone. Traditional methods for accelerating inference, including model compression and migration from language model acceleration, often compromise output quality or face challenges in effectively integrating multimodal features. To address these issues, we propose AASD, a novel framework for Accelerating inference with refined KV Cache and Aligning speculative decoding in MLLMs. Our approach leverages the target model’s cached Key-Value (KV) pairs to extract vital information for generating draft tokens, enabling efficient speculative decoding.
To reduce the computational burden associated with long multimodal token sequences, we introduce a KV Projector to compress the KV Cache while maintaining representational fidelity. Additionally, we design a Target-Draft Attention mechanism that optimizes the alignment between the draft and target models, achieving the benefits of real inference scenarios with minimal computational overhead.
Extensive experiments on mainstream MLLMs demonstrate that our method achieves up to a 2× inference speedup without sacrificing accuracy. This study not only provides an effective and lightweight solution for accelerating MLLM inference but also introduces a novel alignment strategy for speculative decoding in multimodal contexts, laying a strong foundation for future research in efficient MLLMs.
Code is availiable at https://anonymous.4open.science/r/ASD-F571.
Submission Number: 7
Loading