Benchmark Radar
AI BENCHMARK PROFILE

PersonaTrail

General AIAgents

PersonaTrail evaluates personalized web agents using realistic browsing trajectories as user history, assessing preference inference and information recall. Operates in a managed open web environment with two tasks.

Released
2026-05-30
Readiness
Paper only
Primary field
General AI

Why it matters

Addresses the gap in web agent benchmarks by capturing personalization from raw browsing history, moving beyond fully explicit prompts.

Motivation

Recent advances in large language models have enabled web agents to autonomously execute complex tasks.

Primary resources

Benchmark Radar records only publicly supported details and links back to primary sources for verification.