Assessing and Verifying Task Utility in LLM-Powered Applications

TL;DR

This paper introduces AgentEval, a framework that automatically assesses the utility of LLM-powered applications by proposing tailored criteria, ensuring they effectively improve user experience and task efficiency.

Contribution

The paper presents AgentEval, a novel automated framework for verifying the utility of LLM applications through tailored criteria and comprehensive assessment methods.

Findings

AgentEval effectively assesses utility in math problem solving tasks.

AgentEval demonstrates robustness across household-related tasks.

Framework facilitates reproducible utility verification.

Abstract

The rapid development of Large Language Models (LLMs) has led to a surge in applications that facilitate collaboration among multiple agents, assisting humans in their daily tasks. However, a significant gap remains in assessing to what extent LLM-powered applications genuinely enhance user experience and task execution efficiency. This highlights the need to verify utility of LLM-powered applications, particularly by ensuring alignment between the application's functionality and end-user needs. We introduce AgentEval, a novel framework designed to simplify the utility verification process by automatically proposing a set of criteria tailored to the unique purpose of any given application. This allows for a comprehensive assessment, quantifying the utility of an application against the suggested criteria. We present a comprehensive analysis of the effectiveness and robustness of AgentEval for two open source datasets including Math Problem solving and ALFWorld House-hold related tasks. For reproducibility purposes, we make the data, code and all the logs publicly available at https://bit.ly/3w3yKcS .