Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gao, Zhi, Zhang, Bofei, Li, Pengxiang, Ma, Xiaojian, Yuan, Tao, Fan, Yue, Wu, Yuwei, Jia, Yunde, Zhu, Song-Chun, Li, Qing
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915134090248192
author Gao, Zhi
Zhang, Bofei
Li, Pengxiang
Ma, Xiaojian
Yuan, Tao
Fan, Yue
Wu, Yuwei
Jia, Yunde
Zhu, Song-Chun
Li, Qing
author_facet Gao, Zhi
Zhang, Bofei
Li, Pengxiang
Ma, Xiaojian
Yuan, Tao
Fan, Yue
Wu, Yuwei
Jia, Yunde
Zhu, Song-Chun
Li, Qing
contents The advancement of large language models (LLMs) prompts the development of multi-modal agents, which are used as a controller to call external tools, providing a feasible way to solve practical tasks. In this paper, we propose a multi-modal agent tuning method that automatically generates multi-modal tool-usage data and tunes a vision-language model (VLM) as the controller for powerful tool-usage reasoning. To preserve the data quality, we prompt the GPT-4o mini model to generate queries, files, and trajectories, followed by query-file and trajectory verifiers. Based on the data synthesis pipeline, we collect the MM-Traj dataset that contains 20K tasks with trajectories of tool usage. Then, we develop the T3-Agent via \underline{T}rajectory \underline{T}uning on VLMs for \underline{T}ool usage using MM-Traj. Evaluations on the GTA and GAIA benchmarks show that the T3-Agent consistently achieves improvements on two popular VLMs: MiniCPM-V-8.5B and {Qwen2-VL-7B}, which outperforms untrained VLMs by $20\%$, showing the effectiveness of the proposed data synthesis pipeline, leading to high-quality data for tool-usage capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2412_15606
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage
Gao, Zhi
Zhang, Bofei
Li, Pengxiang
Ma, Xiaojian
Yuan, Tao
Fan, Yue
Wu, Yuwei
Jia, Yunde
Zhu, Song-Chun
Li, Qing
Artificial Intelligence
Computer Vision and Pattern Recognition
The advancement of large language models (LLMs) prompts the development of multi-modal agents, which are used as a controller to call external tools, providing a feasible way to solve practical tasks. In this paper, we propose a multi-modal agent tuning method that automatically generates multi-modal tool-usage data and tunes a vision-language model (VLM) as the controller for powerful tool-usage reasoning. To preserve the data quality, we prompt the GPT-4o mini model to generate queries, files, and trajectories, followed by query-file and trajectory verifiers. Based on the data synthesis pipeline, we collect the MM-Traj dataset that contains 20K tasks with trajectories of tool usage. Then, we develop the T3-Agent via \underline{T}rajectory \underline{T}uning on VLMs for \underline{T}ool usage using MM-Traj. Evaluations on the GTA and GAIA benchmarks show that the T3-Agent consistently achieves improvements on two popular VLMs: MiniCPM-V-8.5B and {Qwen2-VL-7B}, which outperforms untrained VLMs by $20\%$, showing the effectiveness of the proposed data synthesis pipeline, leading to high-quality data for tool-usage capabilities.
title Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage
topic Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.15606