arXiv cs.MA· Jingjie Ning, Xueqi Li, Yi Lu, Yibo Kong·· 12 小时前AI 评分31
AgentProofBench：为 Web 智能体评测可复用验证器
AgentProofBench: Benchmarking Reusable Verifiers for Web Agents
AI 导读
AgentProofBench 是面向 Web 智能体的可复用验证器基准，覆盖 97 个任务、79 个任务族和 4 项 Web 服务，将冻结验收规则的配对测试与相同保存开局下的执行连接起来。
正文
Abstract:Reusable Web-agent verification requires a rule to follow changing requirements and support subsequent execution. We introduce AgentProofBench, which connects paired tests of a frozen acceptance rule to execution from identical saved openings. The resource spans 97 tasks, 79 families, and four Web services. Experiments with DeepSeek Flash/Pro establish two linked measurements. Across 48 families, complete-panel and decision-replay tests retain 82/96 Flash and 79/96 Pro source programs; paired reuse-rebuild BA intervals are [-5.47, 1.57] and [-10.16, 0.00] percentage points. Across 22 execution families, deployment-state panel correctness separates program assignments whose two saved openings are both delivered correctly: 32/33 assignments with error-free panels, versus 1/11 with panel errors. All reference-defined program-gate regressions and withheld deliveries occur in the latter group. Neutral continuation attains the highest correct delivery under the tested generic feedback, providing a concrete baseline for measuring the value added by a gate. Released instructions, states, programs, reference rules, and trajectories connect each test failure to its execution consequences. AgentProofBench makes reusable verification assessable at both the requirement and policy levels.
| Subjects: | Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) |
| Cite as: | arXiv:2610.34634 [cs.MA] |
| (or arXiv:2610.34634v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34634 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jingjie Ning [view email]
[v1]
Sun, 25 Oct 2026 23:20:28 UTC (473 KB)
来源：arXiv cs.MA · arxiv.org