Elamur Research⌕
☰
home / papers / 2025 / Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning

Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning

Paper · 2025 · Maggie Huan, Yuetai Li, Tuney Zheng et al.

Summary

The paper asks whether the rapid gains reasoning-tuned LLMs show on math benchmarks such as MATH and AIME reflect broader problem-solving skill or narrow specialization. The authors evaluated more than 20 open-weight reasoning models on three task groups: math, other reasoning (science QA, coding, medical QA, agent planning) and non-reasoning tasks (conversational QA, instruction following, hallucination detection). They introduced a Transferability Index, which measures how much of a model's math improvement carries over to other groups. Across model families and sizes, the training method was the most consistent predictor of transfer. RL-tuned models carried their gains over, while SFT-tuned models often lost general abilities. A controlled experiment on Qwen3-14B-Base, trained on the same math data with either SFT or RL, confirmed this pattern. Analyses of hidden-state PCA, KL divergence and token rank shift showed that SFT moves internal representations and output distributions much further from the base model. The findings call into question post-training recipes that depend heavily on SFT with distilled reasoning traces.

Key points

  • Benchmarks:
    • Math: MATH500, AIME24, AIME25 and OlympiadBench.
    • Other reasoning: GPQA-Diamond, LiveCodeBench, ACPBench and HeadQA.
    • Non-reasoning: CoQA, IFEval, HaluEval and MC-TACO.
  • Transferability Index (TI): per-benchmark gains over the base model are z-normalized, compressed with a signed square root, and weighted toward harder tasks. Each group's score is then expressed as a ratio to the math group's score.
  • Controlled study (UniReason models):
    • Training data came from MATH and DeepScaler.
    • SFT targets were Qwen3-32B chain-of-thought traces, kept only if their final answer was correct (rejection sampling).
    • RL used GRPO with answer correctness as the reward.
  • Results on Qwen3-14B-Base:
    • Math average: base 27.7, RL 53.8, best SFT 49.8.
    • Non-reasoning average: base 45.7, RL 53.2, SFT 21.1 (think) and 29.0 (no-think).
    • TI on non-reasoning tasks: RL +52.2, SFT −104.1 and −278.9.
  • Ablation: the sampling distribution, credit assignment and negative gradients support generalization in RL, while KL regularization has only a subtle effect.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Carnegie Mellon UniversityPittsburgh · United States633 works · 195,218 citations🎓University of WashingtonSeattle · United States422 works · 213,539 citations🎓University of PennsylvaniaPhiladelphia · United States241 works · 66,904 citations🎓Hong Kong Polytechnic UniversityHong Kong · China109 works · 5,346 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 19,503 worksArtificial Intelligence · 18,805 worksArts and Humanities · 17,959 worksLanguage and Linguistics · 17,789 works

🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 4,145 worksReasoning · 1,948 works

Cite

Cite as Maggie Huan, Yuetai Li, Tuney Zheng, Xiaoyu Xu, Seungone Kim, Minxin Du, Radha Poovendran, Graham Neubig and Xiang Yue. 2025. Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning. arXiv:2507.00432.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2507.00432 · arXiv 2507.00432

@misc{huan2025does,
  title = {{Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning}},
  author = {Maggie Huan, Yuetai Li, Tuney Zheng, Xiaoyu Xu, Seungone Kim, Minxin Du, Radha Poovendran, Graham Neubig and Xiang Yue},
  year = {2025},
  doi = {10.48550/arxiv.2507.00432},
  eprint = {2507.00432},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2507.00432}
}

References

69 references, 38 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Jie Chen, Xintian Han, Yu Ma, Xun Zhou, and Liang Xiang. Unlock the correlation between supervised fine-tuning and reinforcement learning in training code large language models. ArXiv preprint, abs/2406.10305, 2024. URL https://arxiv.org/abs/2406.10305. — [arXiv:2406.10305](https://arxiv.org/abs/2406.10305) ↗
  2. Wenhu Chen, Ming Yin, Max Ku, Pan Lu, Yixin Wan, Xueguang Ma, Jianyu Xu, Xinyi Wang, and Tony Xia. TheoremQA: A theorem-driven question answering dataset. In Houda Bouamor, Juan Pino, and Kalika Bali (eds.), Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp.…
  3. Zhoujun Cheng, Shibo Hao, Tianyang Liu, Fan Zhou, Yutao Xie, Feng Yao, Yuexin Bian, Yonghao Zhuang, Nilabjo Dey, Yuheng Zha, Yi Gu, Kun Zhou, Yuqi Wang, Yuan Li, Richard Fan, Jianshu She, Chengqian Gao, Abulhair Saparov, Haonan Li, Taylor W. Killian, Mikhail Yurochkin, Zhengzhong Liu, Eric P. Xing,… find
  4. Tianzhe Chu, Yuexiang Zhai, Jihan Yang, Shengbang Tong, Saining Xie, Dale Schuurmans, Quoc V Le, Sergey Levine, and Yi Ma. Sft memorizes, rl generalizes: A comparative study of foundation model post-training. ArXiv preprint, abs/2501.17161, 2025. URL https://arxiv.org/ abs/2501.17161. ↗
  5. Nouha Dziri, Ximing Lu, Melanie Sclar, Xiang Lorraine Li, Liwei Jian, Bill Yuchen Lin, Peter West, Chandra Bhagavatula, Ronan Le Bras, Jena D. Hwang, Soumya Sanyal, Sean Welleck, Xiang Ren, Allyson Ettinger, Zaı̈d Harchaoui, and Yejin Choi. Faith and fate: Limits of transformers on compositionality.…
  6. Yichen Feng, Zhangchen Xu, Fengqing Jiang, Yuetai Li, Bhaskar Ramasubramanian, Luyao Niu, Bill Yuchen Lin, and Radha Poovendran. Visualsphinx: Large-scale synthetic vision logic puzzles for rl, 2025. URL https://arxiv.org/abs/2505.23977. — [arXiv:2505.23977](https://arxiv.org/abs/2505.23977) ↗
  7. Leo Gao, Jonathan Tow, Baber Abbasi, Stella Biderman, Sid Black, Anthony DiPofi, Charles Foster, Laurence Golding, Jeffrey Hsu, Alain Le Noac’h, Haonan Li, Kyle McDonell, Niklas Muennighoff, Chris Ociepa, Jason Phang, Laria Reynolds, Hailey Schoelkopf, Aviya Skowron, Lintang Sutawika, Eric Tang, Ani… find
  8. Etash Guha, Ryan Marten, Sedrick Keh, Negin Raoof, Georgios Smyrnis, Hritik Bansal, Marianna Nezhurina, Jean Mercat, Trung Vu, Zayne Sprague, et al. Openthoughts: Data recipes for reasoning models. ArXiv preprint, abs/2506.04178, 2025. URL https://arxiv.org/abs/ 2506.04178. ↗
  9. Chaoqun He, Renjie Luo, Yuzhuo Bai, Shengding Hu, Zhen Leng Thai, Junhao Shen, Jinyi Hu, Xu Han, Yujie Huang, Yuxiang Zhang, Jie Liu, Lei Qi, Zhiyuan Liu, and Maosong Sun. Olympiadbench: A challenging benchmark for promoting agi with olympiad-level bilingual multimodal scientific problems. In ACL (1…
  10. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. Measuring massive multitask language understanding. In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021.
  11. OpenReview.net, 2021a. URL https://openreview.net/forum?id=d7KBjmI3GmQ. ↗
  12. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. Measuring mathematical problem solving with the MATH dataset. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2), 2021b. URL h…
  13. Chuxuan Hu, Yuxuan Zhu, Antony Kellermann, Caleb Biddulph, Suppakit Waiwitlikhit, Jason Benn, and Daniel Kang. Breaking barriers: Do reinforcement post training gains transfer to unseen domains?, 2025. URL https://arxiv.org/abs/2506.19733. — [arXiv:2506.19733](https://arxiv.org/abs/2506.19733) ↗
  14. Naman Jain, King Han, Alex Gu, Wen-Ding Li, Fanjia Yan, Tianjun Zhang, Sida Wang, Armando Solar-Lezama, Koushik Sen, and Ion Stoica. Livecodebench: Holistic and contamination free evaluation of large language models for code. In The Thirteenth International Conference on Learning Representations, 20…
  15. Harsha Kokel, Michael Katz, Kavitha Srinivas, and Shirin Sohrabi. Acpbench: Reasoning about action, change, and planning. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pp. 26559–26568, 2025. URL https://ojs.aaai.org/index.php/AAAI/ article/view/34857. ↗
  16. Ananya Kumar, Aditi Raghunathan, Robbie Jones, Tengyu Ma, and Percy Liang. Fine-tuning can distort pretrained features and underperform out-of-distribution, 2022. URL https://arxiv.org/abs/2202.10054. ↗
  17. Nathan Lambert, Jacob Morrison, Valentina Pyatkin, Shengyi Huang, Hamish Ivison, Faeze Brahman, Lester James V. Miranda, Alisa Liu, Nouha Dziri, Shane Lyu, Yuling Gu, Saumya Malik, Victoria Graf, Jena D. Hwang, Jiangjiang Yang, Ronan Le Bras, Oyvind Tafjord, Chris Wilhelm, Luca Soldaini, Noah A. Smi… — Tulu 3: Pushing Frontiers in Open Language Model Post-Training
  18. Junyi Li, Xiaoxue Cheng, Xin Zhao, Jian-Yun Nie, and Ji-Rong Wen. Halueval: A large-scale hallucination evaluation benchmark for large language models. In The 2023 Conference on Empirical Methods in Natural Language Processing, 2023. URL https://openreview.net/ forum?id=bxsrykzSnq. ↗
  19. Xiaomin Li, Zhou Yu, Zhiwei Zhang, Xupeng Chen, Ziji Zhang, Yingying Zhuang, Narayanan Sadagopan, and Anurag Beniwal. When thinking fails: The pitfalls of reasoning for instruction-following in llms, 2025a. URL https://arxiv.org/abs/2505.11423. — [arXiv:2505.11423](https://arxiv.org/abs/2505.11423) ↗
  20. Yuetai Li, Zhangchen Xu, Fengqing Jiang, Bhaskar Ramasubramanian, Luyao Niu, Bill Yuchen Lin, Xiang Yue, and Radha Poovendran. Temporal sampling for forgotten reasoning in llms, 2025b. URL https://arxiv.org/abs/2505.20196. ↗
  21. Yuetai Li, Xiang Yue, Zhangchen Xu, Fengqing Jiang, Luyao Niu, Bill Yuchen Lin, Bhaskar Ramasubramanian, and Radha Poovendran. Small models struggle to learn from strong reasoners. CoRR, abs/2502.12143, February 2025c. URL https://doi.org/10.48550/arXiv. 2502.12143. ↗
  22. Bill Yuchen Lin, Abhilasha Ravichander, Ximing Lu, Nouha Dziri, Melanie Sclar, Khyathi Chandu, Chandra Bhagavatula, and Yejin Choi. The unlocking spell on base llms: Rethinking alignment via in-context learning, 2023. URL https://arxiv.org/abs/2312.01552. ↗
  23. Zeyuan Liu, Ziyu Huan, Xiyao Wang, Jiafei Lyu, Jian Tao, Xiu Li, Furong Huang, and Huazhe Xu. World models with hints of large language models for goal achieving. In Luis Chiruzzo, Alan Ritter, and Lu Wang (eds.), Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Assoc… find
  24. Elita Lobo, Chirag Agarwal, and Himabindu Lakkaraju. On the impact of fine-tuning on chain-of-thought reasoning. ArXiv preprint, abs/2411.15382, 2024. URL https://arxiv.org/abs/ 2411.15382. ↗
  25. Shayne Longpre, Le Hou, Tu Vu, Albert Webson, Hyung Won Chung, Yi Tay, Denny Zhou, Quoc V Le, Barret Zoph, Jason Wei, and Adam Roberts. The flan collection: Designing data and methods for effective instruction tuning. In Andreas Krause, Emma Brunskill, Kyunghyun Cho, Barbara Engelhardt, Sivan Sabato…
  26. Haipeng Luo, Qingfeng Sun, Can Xu, Pu Zhao, Jianguang Lou, Chongyang Tao, Xiubo Geng, Qingwei Lin, Shifeng Chen, and Dongmei Zhang. Wizardmath: Empowering mathematical reasoning for large language models via reinforced evol-instruct. ArXiv preprint, abs/2308.09583, 2023. URL https://arxiv.org/abs/23… ↗
  27. Michael Luo, Sijun Tan, Justin Wong, Xiaoxiang Shi, William Y. Tang, Manan Roongta, Colin Cai, Jeffrey Luo, Li Erran Li, Raluca Ada Popa, and Ion Stoica. Deepscaler: Surpassing o1-preview with a 1.5b model by scaling rl, 2025. URL https://huggingface.co/ agentica-org/DeepScaleR-1.5B-Preview. ↗
  28. Xueguang Ma, Qian Liu, Dongfu Jiang, Ge Zhang, Zejun Ma, and Wenhu Chen. General-reasoner: Advancing llm reasoning across all domains. ArXiv preprint, abs/2505.14652, 2025. URL https://arxiv.org/abs/2505.14652. ↗
  29. Swaroop Mishra, Matthew Finlayson, Pan Lu, Leonard Tang, Sean Welleck, Chitta Baral, Tanmay Rajpurohit, Oyvind Tafjord, Ashish Sabharwal, Peter Clark, and Ashwin Kalyan. LILA: A unified benchmark for mathematical reasoning. In Yoav Goldberg, Zornitsa Kozareva, and Yue Zhang (eds.), Proceedings of th…
  30. Niklas Muennighoff, Zitong Yang, Weijia Shi, Xiang Lisa Li, Li Fei-Fei, Hannaneh Hajishirzi, Luke Zettlemoyer, Percy Liang, Emmanuel Candès, and Tatsunori Hashimoto. s1: Simple test-time scaling, 2025. URL https://arxiv.org/abs/2501.19393. ↗
  31. OpenAI. Learning to reason with llms, 2024. learning-to-reason-with-llms/. find
  32. URL https://openai.com/index/Negin Raoof, Etash Kumar Guha, Ryan Marten, Jean Mercat, Eric Frankel, Sedrick Keh, Hritik Bansal, Georgios Smyrnis, Marianna Nezhurina, Trung Vu, Zayne Rea Sprague, Mike A Merrill, Liangyu Chen, Caroline Choi, Zaid Khan, Sachin Grover, Benjamin Feuer, Ashima Suvarna, Sh… ↗
  33. Siva Reddy, Danqi Chen, and Christopher D. Manning. CoQA: A conversational question answering challenge. Transactions of the Association for Computational Linguistics, 7:249–266, 2019. doi: 10.1162/tacl a 00266. URL https://aclanthology.org/Q19-1016. ↗
  34. David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R. Bowman. GPQA: A graduate-level google-proof q&a benchmark. In First Conference on Language Modeling, 2024. URL https://openreview.net/forum?id=Ti67584b98. ↗
  35. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, Y. K. Li, Y. Wu, and Daya Guo. Deepseekmath: Pushing the limits of mathematical reasoning in open language models, 2024. URL https://arxiv.org/abs/2402. 03300. ↗
  36. Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, and Chuan Wu. Hybridflow: A flexible and efficient rlhf framework. In Proceedings of the Twentieth European Conference on Computer Systems, EuroSys ’25, pp. 1279–1297, New York, NY, USA, 2025. Associ…
  37. Shuqian Sheng, Yi Xu, Tianhang Zhang, Zanwei Shen, Luoyi Fu, Jiaxin Ding, Lei Zhou, Xiaoying Gan, Xinbing Wang, and Chenghu Zhou. RepEval: Effective text evaluation with LLM representation. pp. 7019–7033, November 2024. doi: 10.18653/v1/2024.emnlp-main.398. URL https://aclanthology.org/2024.emnlp-ma… ↗
  38. Yi Su, Dian Yu, Linfeng Song, Juntao Li, Haitao Mi, Zhaopeng Tu, Min Zhang, and Dong Yu. Crossing the reward bridge: Expanding rl with verifiable rewards across diverse domains, 2025. URL https://arxiv.org/abs/2503.23829. — [arXiv:2503.23829](https://arxiv.org/abs/2503.23829) ↗
  39. Yiyou Sun, Shawn Hu, Georgia Zhou, Ken Zheng, Hannaneh Hajishirzi, Nouha Dziri, and Dawn Song. Omega: Can llms reason outside the box in math? evaluating exploratory, compositional, and transformative generalization, 2025. URL https://arxiv.org/abs/2506.18880. ↗
  40. DeepSeek-AI Team. DeepSeek-R1: Incentivizing reasoning capability in llms via reinforcement learning, 2025a. URL https://arxiv.org/abs/2501.12948. ↗
  41. Kimi Team, Angang Du, Bofei Gao, Bowei Xing, Changjiu Jiang, Cheng Chen, Cheng Li, Chenjun Xiao, Chenzhuang Du, Chonghua Liao, Chuning Tang, Congcong Wang, Dehao Zhang, Enming Yuan, Enzhe Lu, Fengxiang Tang, Flood Sung, Guangda Wei, Guokun Lai, Haiqing Guo, Han Zhu, Hao Ding, Hao Hu, Hao Yang, Hao Z… — Kimi k1.5: Scaling Reinforcement Learning with LLMs
  42. Qwen Team. Qwen2.5 technical report, 2025b. URL https://arxiv.org/abs/2412. 15115. ↗
  43. David Vilares and Carlos Gómez-Rodrı́guez. HEAD-QA: A healthcare dataset for complex reasoning. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp. 960–966, Florence, Italy, 2019. Association for Computational Linguistics. doi: 10.18653/v1/P19-1092. URL h…
  44. Peiyi Wang, Lei Li, Zhihong Shao, R. X. Xu, Damai Dai, Yifei Li, Deli Chen, Y. Wu, and Zhifang Sui. Math-shepherd: Verify and reinforce llms step-by-step without human annotations. 2024a. URL https://arxiv.org/abs/2312.08935. ↗
  45. Renxi Wang, Haonan Li, Minghao Wu, Yuxia Wang, Xudong Han, Chiyu Zhang, and Timothy Baldwin. Demystifying instruction mixing for fine-tuning large language models. In ACL (Student Research Workshop), pp. 162–169, 2024b. URL https://aclanthology.org/2024. acl-srw.15. ↗
  46. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V Le, Ed H. Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. Self-consistency improves chain of thought reasoning in language models. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id=… ↗
  47. Yiming Wang, Pei Zhang, Baosong Yang, Derek F. Wong, and Rui Wang. Latent space chain-of-embedding enables output-free LLM self-evaluation. In The Thirteenth International Conference on Learning Representations, 2025. URL https://openreview.net/forum?id= jxo70B9fQo. ↗
  48. Zihan Wang, Yunxuan Li, Yuexin Wu, Liangchen Luo, Le Hou, Hongkun Yu, and Jingbo Shang. Multi-step problem solving through a verifier: An empirical analysis on model-induced process supervision. In Yaser Al-Onaizan, Mohit Bansal, and Yun-Nung Chen (eds.), Findings of the Association for Computationa…
  49. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, brian ichter, Fei Xia, Ed H. Chi, Quoc V Le, and Denny Zhou. Chain of thought prompting elicits reasoning in large language models. In Alice H. Oh, Alekh Agarwal, Danielle Belgrave, and Kyunghyun Cho (eds.), Advances in Neural Information Proce… — Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  50. Tianwen Wei, Jian Luan, Wei Liu, Shuang Dong, and Bin Wang. Cmath: Can your language model pass chinese elementary school math test?, 2023. URL https://arxiv.org/abs/2306. 16636. ↗
  51. Johannes Welbl, Nelson F. Liu, and Matt Gardner. Crowdsourcing multiple choice science questions. In Proceedings of the 3rd Workshop on Noisy User-generated Text, pp. 94–106, Copenhagen, Denmark, 2017. Association for Computational Linguistics. doi: 10.18653/v1/W17-4413. URL https://aclanthology.org… ↗
  52. Annie Wong, Thomas Bäck, Aske Plaat, Niki van Stein, and Anna V. Kononova. Towards a deeper understanding of reasoning capabilities in large language models, 2025. URL https://arxiv.org/abs/2505.10543. — [arXiv:2505.10543](https://arxiv.org/abs/2505.10543) ↗
  53. Tianbao Xie, Fan Zhou, Zhoujun Cheng, Peng Shi, Luoxuan Weng, Yitao Liu, Toh Jing Hua, Junning Zhao, Qian Liu, Che Liu, Zeyu Liu, Yiheng Xu, Hongjin SU, Dongchan Shin, Caiming Xiong, and Tao Yu. Openagents: An open platform for language agents in the wild. In First Conference on Language Modeling, 2…
  54. Xiaoyu Xu, Xiang Yue, Yang Liu, Qingqing Ye, Haibo Hu, and Minxin Du. Unlearning isn’t deletion: Investigating reversibility of machine unlearning in llms, 2025a. URL https://arxiv.org/abs/2505.16831. — [arXiv:2505.16831](https://arxiv.org/abs/2505.16831) ↗
  55. Yi Xu, Bo Xue, Shuqian Sheng, Cheng Deng, Jiaxin Ding, Zanwei Shen, Luoyi Fu, Xinbing Wang, and Chenghu Zhou. Good idea or not, representation of llm could tell. ArXiv preprint, abs/2409.13712, 2024. URL https://arxiv.org/abs/2409.13712. — [arXiv:2409.13712](https://arxiv.org/abs/2409.13712) ↗
  56. Zhangchen Xu, Yuetai Li, Fengqing Jiang, Bhaskar Ramasubramanian, Luyao Niu, Bill Yuchen Lin, and Radha Poovendran. Tinyv: Reducing false negatives in verification improves rl for llm reasoning, 2025b. URL https://arxiv.org/abs/2505.14625. ↗
  57. Junxiao Yang, Jinzhe Tu, Haoran Liu, Xiaoce Wang, Chujie Zheng, Zhexin Zhang, Shiyao Cui, Caishun Chen, Tiantian He, Hongning Wang, Yew-Soon Ong, and Minlie Huang. Barrel: Boundary-aware reasoning for factual and reliable lrms, 2025. URL https://arxiv.org/ abs/2505.13529. — [arXiv:2505.13529](https://arxiv.org/abs/2505.13529) ↗
  58. Edward Yeo, Yuxuan Tong, Morry Niu, Graham Neubig, and Xiang Yue. Demystifying long chain-of-thought reasoning in llms. In ICML 2025, 2025. URL https://arxiv.org/abs/2502. 03373. ↗
  59. Longhui Yu, Weisen Jiang, Han Shi, Jincheng YU, Zhengying Liu, Yu Zhang, James Kwok, Zhenguo Li, Adrian Weller, and Weiyang Liu. Metamath: Bootstrap your own mathematical questions for large language models. In The Twelfth International Conference on Learning Representations, 2024. URL https://openr…
  60. Xiang Yue, Xingwei Qu, Ge Zhang, Yao Fu, Wenhao Huang, Huan Sun, Yu Su, and Wenhu Chen. MAmmoTH: Building math generalist models through hybrid instruction tuning. In The Twelfth International Conference on Learning Representations, 2024a. URL https://openreview.net/forum?id=yLClGs770I. ↗
  61. Xiang Yue, Tianyu Zheng, Ge Zhang, and Wenhu Chen. MAmmoTH2: Scaling instructions from the web. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024b. URL https://openreview.net/forum?id=yVu5dnPlqA. ↗
  62. Weihao Zeng, Yuzhen Huang, Qian Liu, Wei Liu, Keqing He, Zejun Ma, and Junxian He. Simplerl-zoo: Investigating and taming zero reinforcement learning for open base models in the wild, 2025. URL https://arxiv.org/abs/2503.18892. ↗
  63. Yuxiang Zhang, Yuqi Yang, Jiangming Shu, Yuhang Wang, Jinlin Xiao, and Jitao Sang. Openrft: Adapting reasoning foundation model for domain-specific tasks with reinforcement fine-tuning. ArXiv preprint, abs/2412.16849, 2024. URL https://arxiv.org/abs/2412.16849. — [arXiv:2412.16849](https://arxiv.org/abs/2412.16849) ↗
  64. Junhao Zheng, Xidi Cai, Shengjie Qiu, and Qianli Ma. Spurious forgetting in continual learning of language models. In The Thirteenth International Conference on Learning Representations, 2025. URL https://openreview.net/forum?id=ScI7IlKGdI. ↗
  65. Yaowei Zheng, Richong Zhang, Junhao Zhang, Yanhan Ye, and Zheyan Luo. LlamaFactory: Unified efficient fine-tuning of 100+ language models. In Yixin Cao, Yang Feng, and Deyi Xiong (eds.), Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonst…
  66. Ben Zhou, Daniel Khashabi, Qiang Ning, and Dan Roth. “going on a vacation” takes longer than “going for a walk”: A study of temporal commonsense understanding. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natu… find
  67. Jeffrey Zhou, Tianjian Lu, Swaroop Mishra, Siddhartha Brahma, Sujoy Basu, Yi Luan, Denny Zhou, and Le Hou. Instruction-following evaluation for large language models. arXiv preprint arXiv:2311.07911, 2023. URL https://arxiv.org/abs/2311.07911. ↗
  68. Lexin Zhou, Lorenzo Pacchiardi, Fernando Martı́nez-Plumed, Katherine M Collins, Yael Moros-Daval, Seraphina Zhang, Qinlin Zhao, Yitian Huang, Luning Sun, Jonathan E Prunty, et al. General scales unlock ai evaluation with explanatory and predictive power. ArXiv preprint, abs/2503.06378, 2025. URL htt… find
  69. Daniel M Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B Brown, Alec Radford, Dario Amodei, Paul Christiano, and Geoffrey Irving. Fine-tuning language models from human preferences. ArXiv preprint, abs/1909.08593, 2019. URL https://arxiv.org/abs/1909.08593. ↗

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/huan-2025-math-reasoning-improve-general-llm/