PersonaTrail: Benchmarking Personalized Web Agents through Browsing Trails

arXiv cs.AI Papers
benchmark web-agents personalization llm memory browsing

Summary

PersonaTrail is a benchmark for personalized web agents that uses realistic browsing trajectories to evaluate agents' ability to infer user preferences and recall past information. The paper also proposes PACMem, a memory framework that outperforms existing baselines on both tasks.

arXiv:2607.20482v1 Announce Type: new Abstract: Recent advances in large language models have enabled web agents to autonomously execute complex tasks. In practice, users frequently provide underspecified instructions, requiring agents to infer the missing context from their raw browsing histories. Existing benchmarks fail to capture this form of personalization, as they either restrict tasks to fully explicit prompts or abstract web interaction history into simplified forms. To bridge this gap, we introduce PersonaTrail, a benchmark for personalized web agents operating in a managed open web environment. By leveraging realistic browsing trajectories as user history, PersonaTrail evaluates an agent's ability to infer user preferences and recall information from past browsing sessions. We further propose Preference-Aware Contextual Memory (PACMem), a framework that decomposes raw browsing histories into two types of structured memory: factual memories that summarize individual sessions and preference memories that distill recurring behavioral patterns. At inference time, the agent retrieves the most relevant entries from these memories to guide personalized navigation. Extensive experiments show that PACMem consistently outperforms existing memory-based baselines on both tasks.
Original Article
View Cached Full Text

Cached at: 07/24/26, 05:01 AM 

# PersonaTrail: Benchmarking Personalized Web Agents through Browsing Trails
Source: [https://arxiv.org/abs/2607.20482](https://arxiv.org/abs/2607.20482)
[View PDF](https://arxiv.org/pdf/2607.20482)

> Abstract:Recent advances in large language models have enabled web agents to autonomously execute complex tasks\. In practice, users frequently provide underspecified instructions, requiring agents to infer the missing context from their raw browsing histories\. Existing benchmarks fail to capture this form of personalization, as they either restrict tasks to fully explicit prompts or abstract web interaction history into simplified forms\. To bridge this gap, we introduce PersonaTrail, a benchmark for personalized web agents operating in a managed open web environment\. By leveraging realistic browsing trajectories as user history, PersonaTrail evaluates an agent's ability to infer user preferences and recall information from past browsing sessions\. We further propose Preference\-Aware Contextual Memory \(PACMem\), a framework that decomposes raw browsing histories into two types of structured memory: factual memories that summarize individual sessions and preference memories that distill recurring behavioral patterns\. At inference time, the agent retrieves the most relevant entries from these memories to guide personalized navigation\. Extensive experiments show that PACMem consistently outperforms existing memory\-based baselines on both tasks\.

## Submission history

From: Seungbin Yang \[[view email](https://arxiv.org/show-email/8128ca15/2607.20482)\] **\[v1\]**Sat, 30 May 2026 13:27:55 UTC \(13,178 KB\)

Similar Articles