ACL Findings2026
Pause or Fabricate? Training Language Models for Grounded Reasoning
Yiwen Qiu, Linjuan Wu, Yizhou Liu, Yuchen Yan, Jin Ma, Xu Tan, Yao Hu, Daoxin Zhang, Wenqi Zhang, Weiming Lu, Jun Xiao, Yongliang Shen
Abstract
Large language models have achieved remarkable progress on complex reasoning tasks. However, they often implicitly fabricate information when inputs are incomplete, producing confident but unreliable conclusions-a failure mode we term ungrounded reasoning. We argue that this issue arises not from insufficient reasoning capability, but from the lack of inferential boundary awareness-the ability to recognize when the necessary premises for valid inference are missing. To address this issue, we propose Grounded Reasoning via Interactive Reinforcement Learning (GRIL), a multi-turn reinforcement learning framework for grounded reasoning under incomplete information. GRIL decomposes the reasoning process into two stages: clarify and pause, which identifies whether the available information is sufficient, and grounded reasoning, which performs task solving once the necessary premises are established. We design stage-specific rewards to penalize hallucinations, enabling models to detect gaps, stop proactively, and resume reasoning after clarification. Experiments on GSM8K-Insufficient and MetaMATH-Insufficient show that GRIL significantly improves premise detection (up to 45%), leading to a 30% increase in task success while reducing average response length by over 20%. Additional analyses confirm robustness to noisy user responses and generalization to out-of-distribution tasks.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Graph of Thoughts: Solving Elaborate Problems with Large Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Builds on12
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- Inference-Time Intervention: Eliciting Truthful Answers from a Language ModelKenneth Li, Oam Patel, Fernanda B. Viégas, Hanspeter Pfister et al.NeurIPS 2023 · 1,549 citations
- Graph of Thoughts: Solving Elaborate Problems with Large Language ModelsMaciej Besta, Nils Blach, Ales Kubicek, Robert Gerstenberger et al.AAAI 2024 · 1,292 citations
- MetaMath: Bootstrap Your Own Mathematical Questions for Large Language ModelsLonghui Yu, Weisen Jiang, Han Shi, Jincheng Yu et al.ICLR 2024 · 637 citations
- LLMs Get Lost In Multi-Turn ConversationPhilippe Laban, Hiroaki Hayashi, Yingbo Zhou, Jennifer NevilleICLR 2026 · 491 citations
Related papers
- Multimodal Reasoning with Explicit Reasoning Patterns and RewardsHan Qiu, Sheng Jin, Zhongrong Zuo, Ziyue Wang, Qi She, Ling Shao, Shijian LuCVPR Findings 2026
- KnowRL: Exploring Knowledgeable Reinforcement Learning for FactualityBaochang Ren, Shuofei Qiao, Ningyu Zhang, Da Zheng et al.ACL 2026 · 12 citations
- Visual Attention Reasoning via Hierarchical Search and Self-VerificationWei Cai, Jian Zhao, Yuchen Yuan, Tianle Zhang et al.ACL 2026
- Open-World LLM Logical ReasoningYe Mo, Chuan Zhou, Fengxiang Cheng, Jialin Yu et al.ICML 2026
- Learning from Cognition: Enhancing RL Efficiency for LLM Reasoning via Hierarchical Metacognitive Decomposition and RefinementZexu Sun, Yongcheng Zeng, Erxue Min, Heyang Gao et al.ACL 2026
