Metraipolis › archive › dc_3e5fc4b0a804

AgentProofBench: Benchmarking Reusable Verifiers for Web Agents — arXiv:2610.34634

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.34634
License (as archived)
CC0-1.0
Archived
2026-10-27T05:18:51.152Z
SHA-256 of the archived text
0238b9194aba9b3bdd52c630eb87dd732dd74d4399e8cda93b594311d893ae82
Stable URL of this page
https://metraipolis.com/documento/dc_3e5fc4b0a804

How to cite

AgentProofBench: Benchmarking Reusable Verifiers for Web Agents — arXiv:2610.34634 — arXiv (en), https://arxiv.org/abs/2610.34634 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3e5fc4b0a804 (Metraipolis archive dc_3e5fc4b0a804, sha256 0238b9194aba9b3bdd52c630eb87dd732dd74d4399e8cda93b594311d893ae82)
BibTeX
@misc{dc3e5fc4b0a804,
  title = {AgentProofBench: Benchmarking Reusable Verifiers for Web Agents — arXiv:2610.34634},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.34634},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3e5fc4b0a804; sha256 0238b9194aba9b3bdd52c630eb87dd732dd74d4399e8cda93b594311d893ae82},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.34634. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Multiagent Systems (cs.MA) ; Artificial Intelligence (cs.AI); Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2610.34634 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-27T05:17:52.486Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).