arXiv cs.CL· Shuli Jiang, Khushbu Pahwa, Yuting Zhang, Yi Zhang, Wei Xia, Stefano Soatto·· 11 小时前AI 评分31
让 LLM 智能体学会判断何时需要人类决策：基于执行器相对咨询价值的分配方法
Learning When Human Decisions Matter: Executor-Relative Consultation Value for LLM Agents
AI 导读
研究提出“人类决策容量分配”问题：在咨询机会有限、预算受限时，LLM 智能体应在哪些决策上咨询人类。为此提出 Consultation Value Distillation（CVD），用 Monte Carlo 执行器 rollout 离线估计与执行器相关的咨询价值，并蒸馏为轻量预测器，部署时无需再做 rollout 估值。
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.CL · arxiv.org