Beyond Passive Critical Thinking: Fostering Proactive Questioning to Enhance Human-AI Collaboration

TL;DR

This paper introduces proactive critical thinking in AI, where models actively seek missing information to improve reasoning, and presents new benchmarks and RL methods to enhance this capability.

Contribution

It proposes a new paradigm of proactive critical thinking, introduces GSM-MC and GSM-MCE benchmarks, and demonstrates RL techniques to significantly improve models' ability to seek missing information.

Findings

Models struggle with proactive questioning, especially smaller ones.

Reinforcement learning greatly improves proactive critical thinking.

Accuracy on GSM-MC increased from 0.15% to 73.98% with RL.

Abstract

Critical thinking is essential for building robust AI systems, preventing them from blindly accepting flawed data or biased reasoning. However, prior work has primarily focused on passive critical thinking, where models simply reject problematic queries without taking constructive steps to address user requests. In this work, we introduce proactive critical thinking, a paradigm where models actively seek missing or clarifying information from users to resolve their queries better. To evaluate this capability, we present GSM-MC and GSM-MCE, two novel benchmarks based on GSM8K for assessing mathematical reasoning under incomplete or misleading conditions. GSM-MC contains 1,368 math problems with a key variable deliberately removed, requiring models to identify and request the missing information. GSM-MCE further increases the difficulty by introducing irrelevant details to test robustness against distractions. Experiments on Qwen3 and Llama series models show that, while these models excel in traditional reasoning tasks due to extensive post-training and inference-time scaling, they struggle with proactive critical thinking, especially smaller ones. However, we demonstrate that reinforcement learning (RL) can significantly improve this ability. Using our enhanced RL algorithm, we achieve substantial gains, boosting the Qwen3-1.7B's accuracy from 0.15% to 73.98% on GSM-MC. We hope this work advances models that collaborate more effectively with users in problem-solving through proactive critical thinking.