paper-with-me

홈 › Papers

ReasonFlux: Hierarchical LLM Reasoning via Scaling Thought Templates

2025-02-10 · Ling Yang, Zhaochen Yu, Bin Cui, Mengdi Wang

We present that hierarchical LLM reasoning via scaling thought templates can effectively optimize the reasoning search space and outperform the mathematical reasoning capabilities of powerful LLMs like OpenAI o1-preview and DeepSeek V3. We train our ReasonFlux-32B model with only 8 GPUs and introduces three innovations: (i) a structured and generic thought template library, containing around 500 high-level thought templates capable of generalizing to similar or relevant reasoning problems; (ii) performing hierarchical reinforcement learning on a sequence of thought templates instead of long CoTs, optimizing a base LLM to plan out an optimal template trajectory for gradually handling complex problems; (iii) a brand new inference scaling system that enables hierarchical LLM reasoning by adaptively scaling thought templates at inference time. With a template trajectory containing sequential thought templates, our ReasonFlux-32B significantly advances math reasoning capabilities to state-of-the-art levels. Notably, on the MATH benchmark, it achieves an accuracy of 91.2% and surpasses o1-preview by 6.7%. On the USA Math Olympiad (AIME) benchmark, ReasonFlux-32B solves an average of 56.7% of problems, surpassing o1-preview and DeepSeek-V3 by 27% and 45%, respectively. Code: https://github.com/Gen-Verse/ReasonFlux

📄 PDF Abstract BibTeX arXiv:2502.06772

Code (3)

gen-verse/reasonflux 공식 구현 pytorch
yangling0818/buffer-of-thought-llm
yangling0818/supercorrect-llm pytorch

Tasks

Hierarchical Reinforcement LearningLanguage ModelingMathMathematical Reasoning

Methods 이 논문이 사용한 방법론

BASE 설명 없음

Similar Papers 제목 키워드 기반

ReasonFlux-PRM: Trajectory-Aware PRMs for Long Chain-of-Thought Reasoning in LLMs

2025-06-23 · Jiaru Zou, Ling Yang, Jingwen Gu, Jiahao Qiu 외

Process Reward Models (PRMs) have recently emerged as a powerful framework for supervising intermediate reasoning steps in large language models (LLMs). Previous PRMs are primarily trained on model final output responses…

Math

Enhancing Test-Time Scaling of Large Language Models with Hierarchical Retrieval-Augmented MCTS

2025-07-08 · Alex ZH Dou, Zhongwei Wan, Dongfei Cui, Xin Wang 외 arxiv

Test-time scaling has emerged as a promising paradigm in language modeling, leveraging additional computational resources at inference time to enhance model performance. In this work, we introduce R2-LLMs, a novel and ve…

When Thoughts Meet Facts: Reusable Reasoning for Long-Context LMs

2025-10-08 · Soyeong Jeong, Taehee Jung, Sung Ju Hwang, Joo-Kyung Kim 외 arxiv

Recent Long-Context Language Models (LCLMs) can process hundreds of thousands of tokens in a single prompt, enabling new opportunities for knowledge-intensive multi-hop reasoning by integrating large sets of retrieved do…

Co-Evolving LLM Coder and Unit Tester via Reinforcement Learning

2025-06-03 · Yinjie Wang, Ling Yang, Ye Tian, Ke Shen 외

We propose CURE, a novel reinforcement learning framework with a dedicated reward design that co-evolves coding and unit test generation capabilities based on their interaction outcomes, without any ground-truth code as …

Code Generationreinforcement-learningReinforcement Learning

Mixture of Reasonings: Teach Large Language Models to Reason with Adaptive Strategies

2025-07-01 · Tao Xiong, Xavier Hu, Wenyan Fan, Shengyu Zhang arxiv

Large language models (LLMs) excel in complex tasks through advanced prompting techniques like Chain-of-Thought (CoT) and Tree-of-Thought (ToT), but their reliance on manually crafted, task-specific prompts limits adapta…

Prompt Engineering