WebGraphEval: Multi-Turn Trajectory Evaluation for Web Agents using Graph Representation

TL;DR

WebGraphEval introduces a graph-based framework for evaluating web agents that captures structural diversity, identifies key decision points, and offers a comprehensive analysis beyond simple success metrics.

Contribution

It presents a novel graph abstraction for web agent trajectories, enabling multi-path, cross-agent, and efficiency-aware evaluation without modifying existing environments.

Findings

Captures cross-model regularities in web interactions

Highlights redundancy and inefficiency in agent behaviors

Identifies critical decision points overlooked by outcome metrics

Abstract

Current evaluation of web agents largely reduces to binary success metrics or conformity to a single reference trajectory, ignoring the structural diversity present in benchmark datasets. We present WebGraphEval, a framework that abstracts trajectories from multiple agents into a unified, weighted action graph. This representation is directly compatible with benchmarks such as WebArena, leveraging leaderboard runs and newly collected trajectories without modifying environments. The framework canonically encodes actions, merges recurring behaviors, and applies structural analyses including reward propagation and success-weighted edge statistics. Evaluations across thousands of trajectories from six web agents show that the graph abstraction captures cross-model regularities, highlights redundancy and inefficiency, and identifies critical decision points overlooked by outcome-based metrics. By framing web interaction as graph-structured data, WebGraphEval establishes a general methodology for multi-path, cross-agent, and efficiency-aware evaluation of web agents.