EMNLP Findings2025
CogAtom: From Cognitive Atoms to Olympiad-level Mathematical Reasoning in Large Language Models
Zhuofan Chen, Jiyuan He, Yichi Zhang, Xing Hu, Haoxing Wen, Jun Bai, Wenge Rong
Abstract
Mathematical reasoning poses significant challenges for Large Language Models (LLMs) due to its demand for multi-step reasoning and abstract conceptual integration. While recent test-time scaling techniques rely heavily on high-quality, challenging problems, the scarcity of Olympiad-level math problems remains a bottleneck. We introduce CogAtom, a novel cognitive atom-based framework for synthesizing mathematically rigorous and cognitively diverse problems. Unlike prior approaches, CogAtom models problem construction as a process of selecting and recombining fundamental reasoning units, cognitive atoms, extracted from human-authored solutions. A diversity-promoting random walk algorithm enables exploration of the cognitive atom space, while a constraint-based recombination mechanism ensures logical soundness and structural validity. The combinatorial nature of the graph structure provides a near-infinite space of reasoning paths, and the walk algorithm systematically explores this space to achieve large-scale synthesis of high-quality problems; meanwhile, by controlling the number of cognitive atoms, we can precisely adjust problem difficulty, ensuring diversity, scalability, and controllability of the generated problems. Experimental results demonstrate that CogAtom outperforms existing methods in accuracy, reasoning depth, and diversity, generating problems that closely match the difficulty of AIME while exceeding it in structural variation. Our work offers a cognitively grounded pathway toward scalable, high-quality math problem generation. 1
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
- Key-Point-Driven Data Synthesis with Its Enhancement on Mathematical Reasoning cites Augmenting Math Word Problems via Iterative Question Composing
- Key-Point-Driven Data Synthesis with Its Enhancement on Mathematical Reasoning cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Augmenting Math Word Problems via Iterative Question Composing cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Augmenting Math Word Problems via Iterative Question Composing cites MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
- MathScale: Scaling Instruction Tuning for Mathematical Reasoning cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- MathScale: Scaling Instruction Tuning for Mathematical Reasoning cites MetaMath: Bootstrap Your Own Mathematical Questions for 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 on19
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- MetaMath: Bootstrap Your Own Mathematical Questions for Large Language ModelsLonghui Yu, Weisen Jiang, Han Shi, Jincheng Yu et al.ICLR 2024 · 637 citations
- MathScale: Scaling Instruction Tuning for Mathematical ReasoningZhengyang Tang, Xingxing Zhang, Benyou Wang, Furu WeiICML 2024 · 163 citations
- Key-Point-Driven Data Synthesis with Its Enhancement on Mathematical ReasoningYiming Huang, Xiao Liu, Yeyun Gong, Zhibin Gou et al.AAAI 2025 · 74 citations
- Augmenting Math Word Problems via Iterative Question ComposingHaoxiong Liu, Yifan Zhang, Yifan Luo, Andrew C. YaoAAAI 2025 · 74 citations
Related papers
- PromptCoT: Synthesizing Olympiad-level Problems for Mathematical Reasoning in Large Language ModelsXueliang Zhao, Wei Wu, Jian Guan, Lingpeng KongACL Findings 2025 · 4 citations
- MathSmith: Towards Extremely Hard Mathematical Reasoning by Forging Synthetic Problems with a Reinforced PolicyShaoxiong Zhan, Yanlin Lai, Ziyu Lu, Dahua Lin et al.AAAI 2026 · 17 citations
- AMO-Bench: Large Language Models Still Struggle in High School Math CompetitionsJunlin Liu, Shengnan An, Shuang Zhou, Dan Ma et al.ACL Findings 2026
- Scaling Code-Assisted Chain-of-Thoughts and Instructions for Model ReasoningHonglin Lin, Qizhi Pei, Zhuoshi Pan, Yu Li et al.NeurIPS 2025 · 12 citations
- MathNet: A Global Multimodal Benchmark for Mathematical Reasoning and RetrievalShaden Alshammari, Kevin Wen, Abrar Zainal, Mark Hamilton et al.ICLR 2026 · 9 citations
