dorsal/arxiv
View SchemaValue of Information: A Framework for Human-Agent Communication
| Authors | Yijiang River Dong, Tiancheng Hu, Zheng Hui, Caiqi Zhang, Ivan Vulić, Andreea Bobu, Nigel Collier |
|---|---|
| Categories | |
| ArXiv ID | 2601.06407vv1 |
| URL | https://arxiv.org/abs/2601.06407 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Large Language Model (LLM) agents deployed for real-world tasks face a fundamental dilemma: user requests are underspecified, yet agents must decide whether to act on incomplete information or interrupt users for clarification. Existing approaches either rely on brittle confidence thresholds that require task-specific tuning, or fail to account for the varying stakes of different decisions. We introduce a decision-theoretic framework that resolves this trade-off through the Value of Information (VoI), enabling agents to dynamically weigh the expected utility gain from asking questions against the cognitive cost imposed on users. Our inference-time method requires no hyperparameter tuning and adapts seamlessly across contexts-from casual games to medical diagnosis. Experiments across four diverse domains (20 Questions, medical diagnosis, flight booking, and e-commerce) show that VoI consistently matches or exceeds the best manually-tuned baselines, achieving up to 1.36 utility points higher in high-cost settings. This work provides a parameter-free framework for adaptive agent communication that explicitly balances task risk, query ambiguity, and user effort.
{
"annotation_id": "df841a1e-dfb7-423b-b6bc-c98aca0191d7",
"date_created": "2026-02-17T05:53:08.236000Z",
"date_modified": "2026-02-17T05:53:08.236000Z",
"file_hash": "524fc29714ed4155df72c558921df50a34dd5a252b1827e1ea5256562ffad62f",
"private": false,
"record": {
"abstract": "Large Language Model (LLM) agents deployed for real-world tasks face a fundamental dilemma: user requests are underspecified, yet agents must decide whether to act on incomplete information or interrupt users for clarification. Existing approaches either rely on brittle confidence thresholds that require task-specific tuning, or fail to account for the varying stakes of different decisions. We introduce a decision-theoretic framework that resolves this trade-off through the Value of Information (VoI), enabling agents to dynamically weigh the expected utility gain from asking questions against the cognitive cost imposed on users. Our inference-time method requires no hyperparameter tuning and adapts seamlessly across contexts-from casual games to medical diagnosis. Experiments across four diverse domains (20 Questions, medical diagnosis, flight booking, and e-commerce) show that VoI consistently matches or exceeds the best manually-tuned baselines, achieving up to 1.36 utility points higher in high-cost settings. This work provides a parameter-free framework for adaptive agent communication that explicitly balances task risk, query ambiguity, and user effort.",
"arxiv_id": "2601.06407",
"authors": [
"Yijiang River Dong",
"Tiancheng Hu",
"Zheng Hui",
"Caiqi Zhang",
"Ivan Vuli\u0107",
"Andreea Bobu",
"Nigel Collier"
],
"categories": [
"cs.CL"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Value of Information: A Framework for Human-Agent Communication",
"url": "https://arxiv.org/abs/2601.06407",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "bc688a7e-b08f-46d5-ba2c-4263aeaabd6f",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}