arXiv cs.AI· Shuli Jiang, Khushbu Pahwa, Yuting Zhang, Yi Zhang, Wei Xia, Stefano Soatto·· 3 小时前AI 评分29
LLM 智能体如何学会在关键决策时咨询人类：CVD 方法提出
Learning When Human Decisions Matter: Executor-Relative Consultation Value for LLM Agents
AI 导读
arXiv 论文提出 Consultation Value Distillation（CVD）框架，解决 LLM 智能体在有限预算下应何时咨询人类的问题。CVD 用 Monte Carlo executor rollouts 离线估算咨询价值并蒸馏为轻量预测器，部署时无需 rollout 式价值估算。
正文 · 原文
Abstract:Large language models increasingly automate complex tasks, but human decision capacity does not scale with their capabilities. In many applications, humans remain essential for providing private context, authority, or judgments unavailable to the agent. We formulate human decision capacity allocation: given multiple consultation opportunities and a limited budget, at which decisions should an LLM agent consult a human? The central challenge is estimating consultation value, which is executor-dependent because the same human input may benefit different downstream agents differently. We introduce Consultation Value Distillation (CVD), which estimates consultation values offline using Monte Carlo executor rollouts and distills these values into lightweight predictors for budgeted allocation, requiring no rollout-based value estimation at deployment. Across five batch and three sequential settings with two frontier executors, CVD improves allocation through joint and continuation-aware value learning, while independent consultation values remain competitive in several settings. CVD also broadly outperforms direct prompting and Value of Information where applicable, remains robust with substantially less rollout supervision, and enables nearly lossless approximate allocation at low search cost.
| Comments: | 31 pages, 20 tables, 2 figures |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.05814 [cs.CL] |
| (or arXiv:2611.05814v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.05814 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shuli Jiang [view email]
[v1]
Wed, 4 Nov 2026 01:45:28 UTC (255 KB)
来源：arXiv cs.AI · arxiv.org