By My Eyes: Grounding Multimodal Large Language Models with Sensor Data via Visual Prompting

TL;DR

This paper introduces a visual prompting method for multimodal large language models to improve performance and efficiency in sensory data tasks by visualizing sensor data and automating visualization creation.

Contribution

It presents a novel visual prompting approach and an automated visualization generator that enhance LLMs' handling of sensor data without task-specific prior knowledge.

Findings

10% higher accuracy over text prompts

Reduced token costs by 15.8 times

Effective across nine sensory tasks and four modalities

Abstract

Large language models (LLMs) have demonstrated exceptional abilities across various domains. However, utilizing LLMs for ubiquitous sensing applications remains challenging as existing text-prompt methods show significant performance degradation when handling long sensor data sequences. We propose a visual prompting approach for sensor data using multimodal LLMs (MLLMs). We design a visual prompt that directs MLLMs to utilize visualized sensor data alongside the target sensory task descriptions. Additionally, we introduce a visualization generator that automates the creation of optimal visualizations tailored to a given sensory task, eliminating the need for prior task-specific knowledge. We evaluated our approach on nine sensory tasks involving four sensing modalities, achieving an average of 10% higher accuracy than text-based prompts and reducing token costs by 15.8 times. Our findings highlight the effectiveness and cost-efficiency of visual prompts with MLLMs for various sensory tasks. The source code is available at https://github.com/diamond264/ByMyEyes.