EMNLP Findings2025
Prejudge-Before-Think: Enhancing Large Language Models at Test-Time by Process Prejudge Reasoning
Jianing Wang, Jin Jiang, Yang Liu, Mengdi Zhang, Xunliang Cai
Abstract
In this paper, we introduce a new process prejudge strategy in LLM reasoning to demonstrate that bootstrapping with process prejudge allows the LLM to adaptively anticipate the errors encountered when advancing the subsequent reasoning steps, similar to people sometimes pausing to think about what mistakes may occur and how to avoid them, rather than relying solely on trial and error.Specifically, we define a prejudge node in the rationale, which represents a reasoning step, with at least one step that follows the prejudge node that has no paths toward the correct answer.To synthesize the prejudge reasoning process, we present an automated reasoning framework with a dynamic tree-searching strategy.This framework requires only one LLM to perform answer judging, response critiquing, prejudge generation, and thought completion.Furthermore, we develop a two-phase training mechanism with supervised fine-tuning (SFT) and reinforcement learning (RL) to further enhance the reasoning capabilities of LLMs.Experimental results from competition-level complex reasoning demonstrate that our method can teach the model to prejudge before thinking and significantly enhance the reasoning ability of LLMs 1 .1 Code and data is released at https://github.com/ wjn1996/Prejudge-Before-Think.Prejudge in Real World Key Step: The entrance of the curve Prejudge Hint: Warning Road Ahead Note: When driving a car reaching a curve near to the cliff, a warning sign which can be viewed as a prejudge hint to explicitly elicit drivers to slow down to avoid failing Prejudge in LLM Reasoning Key Step: The reasoning step where exists one next step that has no correct rationales Prejudge Hint: The denominator cannot be 0 Note: When LLM finishes step 1, the searching procedure will be performed and find that there exists Step 2 that has no correct rationale, so it should provide prejudge hint after Step 1 Q: If and represents the sum of real when .Please calculate ....
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
- Direct Preference Optimization: Your Language Model is Secretly a Reward Model cites Training language models to follow instructions with human feedback
- Reflexion: language agents with verbal reinforcement learning cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Large Language Models are Zero-Shot Reasoners cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Large Language Models are Zero-Shot Reasoners cites Training language models to follow instructions with human feedback
- Chain-of-Thought Prompting Elicits Reasoning in Large Language Models cites Training language models to follow instructions with human feedback
Builds on26
- Training language models to follow instructions with human feedbackLong Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida et al.NeurIPS 2022 · 24,707 citations
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- Direct Preference Optimization: Your Language Model is Secretly a Reward ModelRafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning et al.NeurIPS 2023 · 10,924 citations
- Large Language Models are Zero-Shot ReasonersTakeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo et al.NeurIPS 2022 · 8,168 citations
- Reflexion: language agents with verbal reinforcement learningNoah Shinn, Federico Cassano, Ashwin Gopinath, Karthik Narasimhan et al.NeurIPS 2023 · 5,828 citations
Related papers
- BRiTE: Bootstrapping Reinforced Thinking Process to Enhance Language Model ReasoningHan Zhong, Yutong Yin, Shenao Zhang, Xiaojun Xu et al.ICML 2025
- GPO: Learning from Critical Steps to Improve LLM ReasoningJiahao Yu, Zelei Cheng, Xian Wu, Xinyu XingNeurIPS 2025 · 10 citations
- d1: Scaling Reasoning in Diffusion Large Language Models via Reinforcement LearningSiyan Zhao, Devaansh Gupta, Qinqing Zheng, Aditya GroverNeurIPS 2025 · 191 citations
- Light-IF: Endowing LLMs with Generalizable Reasoning via Preview and Self-Checking for Complex Instruction FollowingChenyang Wang, Liang Wen, Shousheng Jia, Xiangzheng Zhang et al.AAAI 2026 · 5 citations
- ReFT: Reasoning with Reinforced Fine-TuningLuong Quoc Trung, Xinbo Zhang, Zhanming Jie, Peng Sun et al.ACL 2024
