Persona2Web: Benchmarking Personalized Web Agents for Contextual Reasoning with User History

TL;DR

Persona2Web introduces a comprehensive benchmark for evaluating personalized web agents' ability to interpret ambiguous queries by leveraging user history, addressing a key gap in current web agent capabilities.

Contribution

It presents the first benchmark for assessing personalized web agents on real web data, focusing on inferring user preferences from history without explicit instructions.

Findings

Agents face significant challenges in personalization based on user history.

Performance varies across different architectures and models.

The benchmark enables fine-grained evaluation of personalization capabilities.

Abstract

Large language models have advanced web agents, yet current agents lack personalization capabilities. Since users rarely specify every detail of their intent, practical web agents must be able to interpret ambiguous queries by inferring user preferences and contexts. To address this challenge, we present Persona2Web, the first benchmark for evaluating personalized web agents on the real open web, built upon the clarify-to-personalize principle, which requires agents to resolve ambiguity based on user history rather than relying on explicit instructions. Persona2Web consists of: (1) user histories that reveal preferences implicitly over long time spans, (2) ambiguous queries that require agents to infer implicit user preferences, and (3) a reasoning-aware evaluation framework that enables fine-grained assessment of personalization. We conduct extensive experiments across various agent architectures, backbone models, history access schemes, and queries with varying ambiguity levels, revealing key challenges in personalized web agent behavior. For reproducibility, our codes and datasets are publicly available at https://anonymous.4open.science/r/Persona2Web-73E8.