Metraipolis › archive › dc_80a1920e8a8e

Clarify the User or Verify the World? Uncertainty Routing for Proactive Agents — arXiv:2609.32255

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32255
License (as archived)
CC0-1.0
Archived
2026-10-07T20:55:15.231Z
SHA-256 of the archived text
481d31573d28379391ff198e68fc645a85a4d2e7b8b4cd36cc77a4d32211d3bb
Stable URL of this page
https://metraipolis.com/documento/dc_80a1920e8a8e

How to cite

Clarify the User or Verify the World? Uncertainty Routing for Proactive Agents — arXiv:2609.32255 — arXiv (en), https://arxiv.org/abs/2609.32255 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_80a1920e8a8e (Metraipolis archive dc_80a1920e8a8e, sha256 481d31573d28379391ff198e68fc645a85a4d2e7b8b4cd36cc77a4d32211d3bb)
BibTeX
@misc{dc80a1920e8a8e,
  title = {Clarify the User or Verify the World? Uncertainty Routing for Proactive Agents — arXiv:2609.32255},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32255},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_80a1920e8a8e; sha256 481d31573d28379391ff198e68fc645a85a4d2e7b8b4cd36cc77a4d32211d3bb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32255. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Tool-using LLM agents must decide not only whether additional information is needed, but also which source can resolve the uncertainty. Existing proactive approaches often specialize in either user clarification or environment verification, without explicitly determining the appropriate information source for each decision. We formulate this problem as uncertainty routing among ACT, CLARIFY, and VERIFY, and propose PROUR, a proactive uncertainty routing framework. PROUR decomposes action uncertainty into disagreement across plausible user-goal interpretations, which signals user-side ambiguity, and the entropy remaining within each interpretation, which signals missing world-side evidence. To acquire information from the routed source, a query generator is trained with a mode-conditioned information-gain reward, targeting user-goal identification under CLARIFY and next-action identification under VERIFY. On $\tau$-bench, PROUR achieves 28.17% average success rate across retail and airline, outperforming the strongest prior method by 4.57% while using 2.17 fewer interaction steps. The learned policy further generalizes to stronger task agents and transactional domains of $\tau^3$-bench without retraining, demonstrating the benefit of source-aligned uncertainty resolution for proactive agents.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.32255 — 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 2026-10-07T20:29:00.739Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).