Less is More: A Simple yet Effective Token Reduction Method for Efficient Multi-modal LLMs
This work addresses efficiency issues for users of MLLMs, but it is incremental as it builds on existing methods for token reduction.
The paper tackles the high resource consumption of Multimodal Large Language Models (MLLMs) by introducing TRIM, a token reduction method that reduces computational overhead while maintaining performance, as tested on 12 datasets.
The rapid advancement of Multimodal Large Language Models (MLLMs) has led to remarkable performances across various domains. However, this progress is accompanied by a substantial surge in the resource consumption of these models. We address this pressing issue by introducing a new approach, Token Reduction using CLIP Metric (TRIM), aimed at improving the efficiency of MLLMs without sacrificing their performance. Inspired by human attention patterns in Visual Question Answering (VQA) tasks, TRIM presents a fresh perspective on the selection and reduction of image tokens. The TRIM method has been extensively tested across 12 datasets, and the results demonstrate a significant reduction in computational overhead while maintaining a consistent level of performance. This research marks a critical stride in efficient MLLM development, promoting greater accessibility and sustainability of high-performing models.