MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action

TL;DR

MM-REACT is a system that combines ChatGPT with vision experts using a novel prompt design to enable advanced multimodal reasoning and action in zero-shot scenarios.

Contribution

It introduces a new prompt-based system paradigm that allows language models to process and reason over multimodal visual signals without fine-tuning.

Findings

Effective zero-shot multimodal reasoning demonstrated

Versatile application across different visual understanding scenarios

Comparable or superior to fine-tuning approaches

Abstract

We propose MM-REACT, a system paradigm that integrates ChatGPT with a pool of vision experts to achieve multimodal reasoning and action. In this paper, we define and explore a comprehensive list of advanced vision tasks that are intriguing to solve, but may exceed the capabilities of existing vision and vision-language models. To achieve such advanced visual intelligence, MM-REACT introduces a textual prompt design that can represent text descriptions, textualized spatial coordinates, and aligned file names for dense visual signals such as images and videos. MM-REACT's prompt design allows language models to accept, associate, and process multimodal information, thereby facilitating the synergetic combination of ChatGPT and various vision experts. Zero-shot experiments demonstrate MM-REACT's effectiveness in addressing the specified capabilities of interests and its wide application in different scenarios that require advanced visual understanding. Furthermore, we discuss and compare MM-REACT's system paradigm with an alternative approach that extends language models for multimodal scenarios through joint finetuning. Code, demo, video, and visualization are available at https://multimodal-react.github.io/