WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Drouin, Alexandre, Gasse, Maxime, Caccia, Massimo, Laradji, Issam H., Del Verme, Manuel, Marty, Tom, Boisvert, Léo, Thakkar, Megh, Cappart, Quentin, Vazquez, David, Chapados, Nicolas, Lacoste, Alexandre
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929431234215936
author Drouin, Alexandre
Gasse, Maxime
Caccia, Massimo
Laradji, Issam H.
Del Verme, Manuel
Marty, Tom
Boisvert, Léo
Thakkar, Megh
Cappart, Quentin
Vazquez, David
Chapados, Nicolas
Lacoste, Alexandre
author_facet Drouin, Alexandre
Gasse, Maxime
Caccia, Massimo
Laradji, Issam H.
Del Verme, Manuel
Marty, Tom
Boisvert, Léo
Thakkar, Megh
Cappart, Quentin
Vazquez, David
Chapados, Nicolas
Lacoste, Alexandre
contents We study the use of large language model-based agents for interacting with software via web browsers. Unlike prior work, we focus on measuring the agents' ability to perform tasks that span the typical daily work of knowledge workers utilizing enterprise software systems. To this end, we propose WorkArena, a remote-hosted benchmark of 33 tasks based on the widely-used ServiceNow platform. We also introduce BrowserGym, an environment for the design and evaluation of such agents, offering a rich set of actions as well as multimodal observations. Our empirical evaluation reveals that while current agents show promise on WorkArena, there remains a considerable gap towards achieving full task automation. Notably, our analysis uncovers a significant performance disparity between open and closed-source LLMs, highlighting a critical area for future exploration and development in the field.
format Preprint
id arxiv_https___arxiv_org_abs_2403_07718
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?
Drouin, Alexandre
Gasse, Maxime
Caccia, Massimo
Laradji, Issam H.
Del Verme, Manuel
Marty, Tom
Boisvert, Léo
Thakkar, Megh
Cappart, Quentin
Vazquez, David
Chapados, Nicolas
Lacoste, Alexandre
Machine Learning
Artificial Intelligence
We study the use of large language model-based agents for interacting with software via web browsers. Unlike prior work, we focus on measuring the agents' ability to perform tasks that span the typical daily work of knowledge workers utilizing enterprise software systems. To this end, we propose WorkArena, a remote-hosted benchmark of 33 tasks based on the widely-used ServiceNow platform. We also introduce BrowserGym, an environment for the design and evaluation of such agents, offering a rich set of actions as well as multimodal observations. Our empirical evaluation reveals that while current agents show promise on WorkArena, there remains a considerable gap towards achieving full task automation. Notably, our analysis uncovers a significant performance disparity between open and closed-source LLMs, highlighting a critical area for future exploration and development in the field.
title WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2403.07718