WebSuite: Systematically Evaluating Why Web Agents Fail

TL;DR

WebSuite is a diagnostic benchmark that systematically evaluates why web agents fail by breaking down their actions, enabling targeted improvements and revealing specific weaknesses in different agent types.

Contribution

The paper introduces WebSuite, the first extensible benchmark with a taxonomy of web actions to diagnose failure points in generalist web agents.

Findings

Identified distinct failure patterns in web agents.

Demonstrated the utility of WebSuite in pinpointing weaknesses.

Provided insights for targeted agent improvements.

Abstract

We describe WebSuite, the first diagnostic benchmark for generalist web agents, designed to systematically evaluate why agents fail. Advances in AI have led to the rise of numerous web agents that autonomously operate a browser to complete tasks. However, most existing benchmarks focus on strictly measuring whether an agent can or cannot complete a task, without giving insight on why. In this paper, we 1) develop a taxonomy of web actions to facilitate identifying common failure patterns, and 2) create an extensible benchmark suite to assess agents' performance on our taxonomized actions. This benchmark suite consists of both individual tasks, such as clicking a button, and end-to-end tasks, such as adding an item to a cart, and is designed such that any failure of a task can be attributed directly to a failure of a specific web action. We evaluate two popular generalist web agents, one text-based and one multimodal, and identify unique weaknesses for each agent. Because WebSuite can disaggregate task failures into specific action failures, this enables granular identification of which UX flows an individual agent has trouble with and immediately highlights promising avenues for improvement. These findings highlight the need for more focused benchmarking on where web agents go wrong to effectively improve agents beyond their weaker performance today.