When and How Should an Agent Clarify? CIGAsk: Teaching LLMs to Clarify via Counterfactual Information Gain

arXiv:2609.24290 · cs.AI · Submitted 2026-09-21 · Read on arXiv

cs.AI

Submitted: 2026-09-21

Updated: 2026-09-21

Comments: Accepted to EMNLP 2026 (Findings)

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• GRPO — 21× in this paper · explained in 6 episodes
• Benchmark — 18× in this paper · explained in 2 episodes
• Information Gain — 11× in this paper · explained in 2 episodes
• Actor — 9× in this paper · explained in 3 episodes
• Counterfactual Information Gain — 8× in this paper
• Ablation — 5× in this paper · explained in 1 episode
• Reinforcement Learning — 5× in this paper · explained in 4 episodes
• Credit assignment — 3× in this paper · explained in 3 episodes
• Adam — 2× in this paper · explained in 2 episodes
• Degeneracy — 2× in this paper · explained in 3 episodes
• Verifier — 2× in this paper · explained in 2 episodes
• Accuracy — 1× in this paper · explained in 1 episode
• Active Learning — 1× in this paper · explained in 6 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Decode — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Frozen model — 1× in this paper · explained in 3 episodes
• Grounding — 1× in this paper · explained in 3 episodes
• Knowledge Base Construction — 1× in this paper · explained in 2 episodes
• Objective Function — 1× in this paper · explained in 2 episodes
• Reward Shaping — 1× in this paper · explained in 2 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Self-consistency — 1× in this paper · explained in 2 episodes

Sources

• ConvAI3: Generating Clarifying Questions for Open-Domain Dialogue Systems (ClariQ)
• STaR-GATE: Teaching Language Models to Ask Clarifying Questions
• Learning Steerable Clarification Policies with Collaborative Self-play
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• InfoPO: Information-Driven Policy Optimization for User-Centric Agents
• AmbigDocs: Reasoning across Documents on Different Entities under the Same Name
• QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?
• Agentic Reinforcement Learning with Implicit Step Rewards
• UserRL: Training Interactive User-Centric Agent via Reinforcement Learning
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• GRPO is Secretly a Process Reward Model
• Structured Uncertainty guided Clarification for LLM Agents
• Information Gain-based Policy Optimization: A Simple and Effective Approach for Multi-Turn Search Agents
• Reinforcing Multi-Turn Reasoning in LLM Agents via Turn-Level Reward Design
• Qwen2.5 Technical Report
• DAPO: An Open-Source LLM Reinforcement Learning System at Scale
• SWEET-RL: Training Multi-Turn LLM Agents on Collaborative Reasoning Tasks
• When and What to Ask: AskBench and Rubric-Guided RLVR for LLM Clarification

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection