Clarify the User or Verify the World? Uncertainty Routing for Proactive Agents

Zhaofeng Li; Xuan Zhang; Xiaokui Xiao; Yang Deng

cs.AISep 26, 2026

Clarify the User or Verify the World? Uncertainty Routing for Proactive Agents

Authors: Zhaofeng Li, Xuan Zhang, Xiaokui Xiao, Yang Deng

Organizations: National University of Singapore · Singapore Management University

Abstract

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 $τ$ -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 $τ^3$ -bench without retraining, demonstrating the benefit of source-aligned uncertainty resolution for proactive agents.

Figures & tables

Figure 1: An example of clarification and verification in $\tau$ -airline trajectories. Clarification alone burdens the user with facts available from tools (left), while verification alone misses user preferences (middle). Combining them resolves the preference before retrieving system evidence (right).

Figure 2: Source distribution of 19,022 step-level events from $\tau$ -bench. World-side cases (blue) account for 89.1%; user-side cases (orange) account for 10.9%. See Appendix A.1 for the full taxonomy.

Panel of Figure 3 — Figure 3: Overview of the Prour framework . (a) Decomposer estimates action-level aleatoric and epistemic uncertainty from candidate user-goal interpretations and routes to clarify , verify , or act . (b) Query Generator is trained with a mode-conditioned information-gain reward, using the ground-truth user goal as the clarification target and reference next action as the verification target.

Panel of Figure 4 — Figure 4: Training dynamics under mode-conditioned information-gain reward.

Panel of Figure 5 — Figure 5: Effect of the decomposer-call budget on performance and cost. (a) Success rate on retail, airline, and their average. (b) Average wall-clock time and total interaction steps per episode. The $\infty$ setting imposes no per-episode limit on decomposer calls.

Figure 6: Uncertainty trajectories over interaction. Prour reports AU, EU, and total uncertainty over the first 16 steps, while baselines report total uncertainty with decomposer applied post hoc to trajectories.

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Panel of Figure 7 — Figure 7: Held-out routing performance of our calibrated Prour and the prompted LLM Router. Rows denote ground-truth labels and columns denote predicted modes; each cell reports the count and row-normalized percentage. Prour applies the Qwen3-8B thresholds selected on the retail development set, whereas the prompted Qwen3-8B router predicts a mode directly, without calibrated thresholds. Both methods are evaluated on the same disjoint airline test set.

Metric	Qwen3-8B	GLM-4-32B
$(\tau_{\mathrm{AU}},\tau_{\mathrm{EU}})$	$(0.25,0.20)$	$(0.20,0.08)$
Macro-F1	0.714	0.711
act F1	0.700	0.742
clarify F1	0.683	0.706
verify F1	0.760	0.685

Appendix

Table 6: Routing F1 after model-specific calibration. Macro-F1 and per-class F1 evaluate three-way routing on the same development set.

Panel of Figure 8 — Figure 8: Effect of decomposer invocation frequency on performance and cost. (a) $\mathrm{pass}$ @1 on retail, airline, and their average. (b) Average wall-clock time and total interaction steps per episode. Each x-axis value $n$ denotes invoking the decomposer once every $n$ agent decision steps.

Explore similar work

Uncertainty Decomposition for Clarification Seeking in LLM Agents
Jun 17, 2026Gregory MatsnevLLM Agents LLM Uncertainty Estimation
AI Talent Hub, ITMO University Saint Petersburg 197101, Russia
Uncertainty-Aware Clarification in LLM Agents with Information Gain
Jun 2, 2026Mengyi Deng, Zhiwei Li, Xin Li +4Clarification Question Generation Tool-Using Agents
Information Hub, HKUST(GZ) · Jilin University · et al.
From Sequence to Structure: Relational Uncertainty Propagation for LLM Agents
Aug 17, 2026Zhengzhao Ma, Boxi Cao, Yaojie Lu +3LLM Uncertainty Estimation LLM Agent Reliability
Chinese Information Processing Laboratory, Institute of Software, Chinese Academy of Sciences · University of Chinese Academy of Sciences, Beijing, China

Jun 17, 2026, cs.AIJ/K move · Enter open · S save

Uncertainty Decomposition for Clarification Seeking in LLM Agents

Gregory Matsnev

AI Talent Hub, ITMO University Saint Petersburg 197101, Russia

LLM Agents LLM Uncertainty Estimation Uncertainty Decomposition LLM Agent Reliability

Open paper AskPDF arXiv

Recent position papers argue that the classical aleatoric/epistemic uncertainty framework is insufficient for interactive large language model (LLM) agents and call for underspecification-aware, decomposed, and communicable uncertainty representations that can unlock new agent capabilities such as proactive clarification seeking and shared mental-model building. Practical deployment constraints -- black-box APIs, interactive latency budgets, and the absence of labeled trajectories -- rule out logprob-based, multi-sampling, and training-based methods, leaving prompt-based estimation as the most viable family for surfacing such signals at deployment time. We answer this call with a simple prompt-based decomposition that separates action confidence from request uncertainty (u), enabling the agent to ask for clarification when the task specification is ambiguous. To evaluate it, we introduce two clarification-augmented benchmarks (WebShop-Clarification and ALFWorld-Clarification) in which 50% of tasks are deliberately underspecified, and systematically compare the proposed decomposition against ReAct+UE and Uncertainty-Aware Memory (UAM) across five LLM backbones (GPT-5.1, DeepSeek-v3.2-exp, GLM-4.7, Qwen3.5-35B, GPT-OSS-120B) on these variants together with the standard WebShop, ALFWorld, and REAL benchmarks for fault detection. Averaged across the five backbones, the proposed decomposition improves clarification F1 on ALFWorld-Clarification by 73% over ReAct+UE and by 36% over UAM, and leads clarification F1 on every backbone on WebShop-Clarification and on four of five backbones on ALFWorld-Clarification, indicating that the gains generalize beyond a single LLM.

Abstract

Explore similar work

Uncertainty Decomposition for Clarification Seeking in LLM Agents

Uncertainty-Aware Clarification in LLM Agents with Information Gain

From Sequence to Structure: Relational Uncertainty Propagation for LLM Agents