Metraipolis › archive › dc_32eb4f026c6c
Learning to Clarify Underspecified Intents Under Limited Interaction — arXiv:2610.04719
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.04719
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:24:45.078Z
- SHA-256 of the archived text
1b298312fd346e2d66699263e4f94708581f6ede64ab6014c26d3b56f2e7dbdb- Stable URL of this page
https://metraipolis.com/documento/dc_32eb4f026c6c
How to cite
Learning to Clarify Underspecified Intents Under Limited Interaction — arXiv:2610.04719 — arXiv, https://arxiv.org/abs/2610.04719 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_32eb4f026c6c (Metraipolis archive dc_32eb4f026c6c, sha256 1b298312fd346e2d66699263e4f94708581f6ede64ab6014c26d3b56f2e7dbdb)
BibTeX
@misc{dc32eb4f026c6c,
title = {Learning to Clarify Underspecified Intents Under Limited Interaction — arXiv:2610.04719},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.04719},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_32eb4f026c6c; sha256 1b298312fd346e2d66699263e4f94708581f6ede64ab6014c26d3b56f2e7dbdb},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04719. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
AI assistants receive requests that leave out information needed for a good outcome, for example about users' preferences or goals. They must then either speculate or ask for more information before proceeding. We reconceptualize this as a value-of-information problem: the assistant should acquire information whose absence causes the greatest avoidable loss in user utility. This is rarely known ex ante; rather, assistants must predict it in order to optimally allocate limited user interactions. We instantiate this problem in image generation and derive a reinforcement learning framework using multi-turn simulated users to maximize utility recovery under uncertainty. In a preregistered study with 456 interactive sessions across 76 human participants, this helped users significantly better match reference images with significantly fewer questions, less total interaction time, and lower cost. This points toward a simple and scalable framework for training language model assistants to better disambiguate user intent by asking more informative questions. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.04719 — 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 2026-10-06T05:24:05.506Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).