arXiv：cs.AI· Zhi Li, Tao Zhou, Yeqing Li, Eugene Ie, Demetri Terzopoulos·· 6 小时前AI 评分49
WebRider：面向实时网页辅助的 persona 条件化意图控制器
WebRider: Persona-Conditioned Intent Controllers for Live-Web Assistance
AI 导读
WebRider 提出将网页任务委托形式化为「意图契约」，记录目标、约束、证据义务、答案形式与 persona 控制，以解决实时网页智能体只看最终答案、忽视策略约束的问题——审计显示强控制器任务完成率 99.2%，但完全遵守策略约束仅 38.8%。
正文
Abstract:Delegating a web task involves more than asking a question; it requires transferring a policy: what to verify, how to handle uncertainty, which preferences matter, and when to stop. Yet, current live-web agents are evaluated solely on the final answer, ignoring the policy constraints that define the delegation. A plausible final answer can conceal violations of that policy. Our full live audit reveals this critical gap: a strong controller completes 99.2% of tasks but honors all policy constraints in only 38.8% of cases. Finishing does not imply fidelity. WebRider bridges this gap by formalizing the delegated policy as an intent contract---an operational record of goals, constraints, evidence obligations, answer form, and task-local persona controls that must hold even as web pages change. WebRider employs a hierarchical architecture: a top-layer controller maintains the contract, a middle layer realizes intentions as guarded executable actions, and a tool layer executes these actions via browser, search, and maps tools. Our benchmark, RiderBench, evaluates this design on 4,096 live-web contracts across 42 public websites, auditing both the internal contract state and the visible user experience to determine if a rollout preserved its policy and if the steps were persona-consistent. The guarded middle interface also serves as a high-quality training signal; an 8B action-policy model trained through this interface outperforms executable-only baselines under a fixed controller. By making the browsing path a first-class object, WebRider enables a system that is auditable, human-judgeable, and learnable without conflating action realization with final-answer decisions. Dataset URL: this http URL.
| Subjects: | Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC) |
| Cite as: | arXiv:2608.06704 [cs.AI] |
| (or arXiv:2608.06704v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2608.06704 arXiv-issued DOI via DataCite |
Submission history
From: Zhi Li [view email]
[v1]
Fri, 7 Aug 2026 02:00:49 UTC (5,693 KB)
[v2]
Tue, 6 Oct 2026 00:15:51 UTC (5,846 KB)
来源：arXiv：cs.AI · arxiv.org