Saved in:
Bibliographic Details
Main Authors: Chan, Chi-Min, Yu, Jianxuan, Chen, Weize, Jiang, Chunyang, Liu, Xinyu, Shi, Weijie, Liu, Zhiyuan, Xue, Wei, Guo, Yike
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.14972
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916371196018688
author Chan, Chi-Min
Yu, Jianxuan
Chen, Weize
Jiang, Chunyang
Liu, Xinyu
Shi, Weijie
Liu, Zhiyuan
Xue, Wei
Guo, Yike
author_facet Chan, Chi-Min
Yu, Jianxuan
Chen, Weize
Jiang, Chunyang
Liu, Xinyu
Shi, Weijie
Liu, Zhiyuan
Xue, Wei
Guo, Yike
contents The rapid advancement of large language models (LLMs) has led to the rise of LLM-based agents. Recent research shows that multi-agent systems (MAS), where each agent plays a specific role, can outperform individual LLMs. However, configuring an MAS for a task remains challenging, with performance only observable post-execution. Inspired by scaling laws in LLM development, we investigate whether MAS performance can be predicted beforehand. We introduce AgentMonitor, a framework that integrates at the agent level to capture inputs and outputs, transforming them into statistics for training a regression model to predict task performance. Additionally, it can further apply real-time corrections to address security risks posed by malicious agents, mitigating negative impacts and enhancing MAS security. Experiments demonstrate that an XGBoost model achieves a Spearman correlation of 0.89 in-domain and 0.58 in more challenging scenarios. Furthermore, using AgentMonitor reduces harmful content by 6.2% and increases helpful content by 1.8% on average, enhancing safety and reliability. Code is available at \url{https://github.com/chanchimin/AgentMonitor}.
format Preprint
id arxiv_https___arxiv_org_abs_2408_14972
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle AgentMonitor: A Plug-and-Play Framework for Predictive and Secure Multi-Agent Systems
Chan, Chi-Min
Yu, Jianxuan
Chen, Weize
Jiang, Chunyang
Liu, Xinyu
Shi, Weijie
Liu, Zhiyuan
Xue, Wei
Guo, Yike
Computation and Language
The rapid advancement of large language models (LLMs) has led to the rise of LLM-based agents. Recent research shows that multi-agent systems (MAS), where each agent plays a specific role, can outperform individual LLMs. However, configuring an MAS for a task remains challenging, with performance only observable post-execution. Inspired by scaling laws in LLM development, we investigate whether MAS performance can be predicted beforehand. We introduce AgentMonitor, a framework that integrates at the agent level to capture inputs and outputs, transforming them into statistics for training a regression model to predict task performance. Additionally, it can further apply real-time corrections to address security risks posed by malicious agents, mitigating negative impacts and enhancing MAS security. Experiments demonstrate that an XGBoost model achieves a Spearman correlation of 0.89 in-domain and 0.58 in more challenging scenarios. Furthermore, using AgentMonitor reduces harmful content by 6.2% and increases helpful content by 1.8% on average, enhancing safety and reliability. Code is available at \url{https://github.com/chanchimin/AgentMonitor}.
title AgentMonitor: A Plug-and-Play Framework for Predictive and Secure Multi-Agent Systems
topic Computation and Language
url https://arxiv.org/abs/2408.14972