BOLAA: BENCHMARKING AND ORCHESTRATING LLM AUTONOMOUS AGENTS

Zhiwei Liu, Weiran Yao, Jianguo Zhang, Le Xue, Shelby Heinecke, Rithesh R N, Yihao Feng, Zeyuan Chen, Juan Carlos Niebles, Devansh Arpit, Ran Xu, Phil L Mui, Huan Wang, Caiming Xiong, Silvio Savarese

Keywords: Agent, LLM, Benchmark, Orchestration

TL;DR: The agent benchmark and a new agent orchestration architecture

Abstract:

The massive successes of large language models (LLMs) encourage the emerging exploration of LLM-augmented Autonomous Agents (LAAs).
An LAA is able to generate actions with its core LLM and interact with environments, which facilitates the ability to resolve complex tasks by conditioning on past interactions such as observations and actions.
Since the investigation of LAA is still very recent, limited explorations are available.
Therefore, we provide a comprehensive comparison of LAA in terms of both agent architectures and LLM backbones.
Additionally, we propose a new strategy to orchestrate multiple agents such that each agent focuses on one specific tasks, i.e. BOLAA, where a controller manages the communication among multiple agents.
We conduct simulations on both decision-making and multi-step reasoning environments, which comprehensively justify the capacity of LAAs.
Our performance results provide quantitative suggestions for designing LAA architectures and the optimal choice of LLMs, as well as the compatibility of both.

Submission Number: 82