← 研踪 PaperTrailarXiv:2606.18671 · v2
HANSEL: Interactive Evidence for Post-Hoc Verification of Web Agents
首次提交 2026-06-17 12:13
版本更新 2026-11-09 12:41
雷达观测 2026-11-10 13:18
摘要
AI web agents can perform complex tasks, such as searching for products and making purchases on users' behalf. However, verifying an agent's output remains difficult, because existing transparency mechanisms (e.g., full trajectory logs) require users to review logs and explanations, keeping verification a passive reading task. In this work, we view agent verification as an interactive activity and present HANSEL. Given an agent trajectory, HANSEL extracts evidence pages and snippets and presents them as navigable, live views that preserve relevant page state (e.g., applied filters), allowing users to verify how the agent arrived at its answer through interaction. In a controlled user study with 14 participants, we show that HANSEL significantly reduces task completion time and perceived effort, demonstrating that reframing verification as an interactive activity enables more efficient post-hoc verification of AI agents.
作者备注：26 pages, 7 figures
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-09 12:41 · 最新观测版本
HANSEL: Interactive Evidence for Post-Hoc Verification of Web Agents
AI web agents can perform complex tasks, such as searching for products and making purchases on users' behalf. However, verifying an agent's output remains difficult, because existing transparency mechanisms (e.g., full trajectory logs) require users to review logs and explanations, keeping verification a passive reading task. In this work, we view agent verification as an interactive activity and present HANSEL. Given an agent trajectory, HANSEL extracts evidence pages and snippets and presents them as navigable, live views that preserve relevant page state (e.g., applied filters), allowing users to verify how the agent arrived at its answer through interaction. In a controlled user study with 14 participants, we show that HANSEL significantly reduces task completion time and perceived effort, demonstrating that reframing verification as an interactive activity enables more efficient post-hoc verification of AI agents.
查看此版本解读 →查看此版本原文 ↗