--- title: "General-Reasoner: Advancing LLM Reasoning Across All Domains" description: "Reinforcement learning with verifiable rewards has boosted LLM reasoning, but mostly in math and… 32 references, cited by 42 papers here." authors: ["Xueguang Ma", "Qian Liu", "Dongfu Jiang", "Ge Zhang", "Zejun Ma", "Wenhu Chen"] year: 2025 venue: "arXiv:2505.14652" source: "https://arxiv.org/abs/2505.14652" doi: "10.48550/arxiv.2505.14652" arxiv: "2505.14652" licence: "CC BY 4.0 (arXiv)" licence_url: "http://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/ma-2025-general-reasoner-advancing-llm-reasoning/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/ma-2025-general-reasoner-advancing-llm-reasoning/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/3/) / General-Reasoner: Advancing LLM Reasoning Across All Domains # General-Reasoner: Advancing LLM Reasoning Across All Domains **Authors:** [Xueguang Ma](https://md.research.elamur.ai/free/en/authors/xueguang-ma/), [Qian Liu](https://md.research.elamur.ai/free/en/authors/qian-liu/), [Dongfu Jiang](https://md.research.elamur.ai/free/en/authors/dongfu-jiang/), [Ge Zhang](https://md.research.elamur.ai/free/en/authors/ge-zhang/), [Zejun Ma](https://md.research.elamur.ai/free/en/authors/zejun-ma/) and [Wenhu Chen](https://md.research.elamur.ai/free/en/authors/wenhu-chen/) · **Year:** 2025 · **Venue:** [arXiv:2505.14652](https://md.research.elamur.ai/free/en/venues/neural-information-processing-systems/) ([NeurIPS 2025](https://md.research.elamur.ai/free/en/venues/neural-information-processing-systems/2025/)) · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://arxiv.org/abs/2505.14652) · [PDF](https://arxiv.org/pdf/2505.14652) · [DOI 10.48550/arxiv.2505.14652](https://doi.org/10.48550/arxiv.2505.14652) · [arXiv 2505.14652](https://arxiv.org/abs/2505.14652) · [OpenAlex](https://openalex.org/W4417298821) **Citations:** 1 citations (OpenAlex, 2026-10-14) · [42](https://md.research.elamur.ai/free/en/papers/ma-2025-general-reasoner-advancing-llm-reasoning/cited-by/) in our library, of the 149,439 papers whose bibliographies we read ## Summary Reinforcement learning with verifiable rewards has boosted LLM reasoning, but mostly in math and code. Those areas have plentiful training data and answers that simple rules can check. Other fields such as physics, finance or the humanities have scarcer data and answers in many forms, so rule-based rewards often fail there. General-Reasoner tackles both problems. The authors traced WebInstruct entries back to their source pages to recover human-written answers. They then used Gemini models to keep questions with short checkable answers, tag them and drop items that are trivial or unsolvable. The result is WebInstruct-verified, about 230K multi-domain questions. They also trained General-Verifier, a small generative model that reasons step by step to judge whether a predicted answer matches the reference. Base models trained this way in the Zero-RL setting with GRPO gain roughly 10% on general benchmarks like MMLU-Pro and SuperGPQA, while matching or slightly beating math-focused RL on math. This shows that broad-domain RL is practical without an SFT stage. ## Key points - **Dataset pipeline:** about 5M WebInstruct entries were re-crawled and reduced to about 1M verifiable questions using Gemini-1.5-Pro. Gemini-2.0-Flash then tagged them and generated 8 candidate solutions per question for filtering, leaving about 230K items (WebInstruct-verified). Math below university level was removed for balance. - **General-Verifier (1.5B parameters):** on 50k answers that Gemini-2.0-Flash judged correct, it agreed 78.7% of the time, versus 22.2% for the rule-based Math-Verify. The biggest gaps appear for list and string answers and in Physics and Engineering. - **Evaluation:** 12 benchmarks, including MMLU-Pro, GPQA, SuperGPQA, TheoremQA, BBEH, MATH-500, GSM8K and Olympiad. General-Reasoner-Qw3-14B matches or beats GPT-4o on several of them. - **Ablations:** with Qwen3-4B-Base, the model-based verifier reached 60.1 on MMLU-Pro versus 58.1 for rule-based checking, which plateaued around step 60. Training Qwen2.5-14B-Base on the full data instead of math only raised SuperGPQA from 35.6 to 39.5. - **Efficiency:** responses average about 1,000 tokens. General-Reasoner-4B scores 61% on the MMLU-Pro CS split in 1.5 minutes, while DeepScaleR-1.5B-Preview takes 18 minutes to reach 35%. *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 - [University of Waterloo](https://md.research.elamur.ai/free/en/universities/university-of-waterloo/) — University, Waterloo, Canada · 80 works · 15,030 citations - [University of Michigan](https://md.research.elamur.ai/free/en/universities/university-of-michigan/) — University, Ann Arbor, United States · 86 works · 15,493 citations 📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works 🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works · [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,685 works ## Cite Cite as: Xueguang Ma, Qian Liu, Dongfu Jiang, Ge Zhang, Zejun Ma and Wenhu Chen. 2025. General-Reasoner: Advancing LLM Reasoning Across All Domains. arXiv:2505.14652. The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @misc{ma2025general, title = {{General-Reasoner: Advancing LLM Reasoning Across All Domains}}, author = {Xueguang Ma, Qian Liu, Dongfu Jiang, Ge Zhang, Zejun Ma and Wenhu Chen}, year = {2025}, doi = {10.48550/arxiv.2505.14652}, eprint = {2505.14652}, archivePrefix = {arXiv}, url = {https://doi.org/10.48550/arxiv.2505.14652} } ``` - [← previous Stop Looking for “Important Tokens” in Multimodal Language Models: Duplication Matters More](https://md.research.elamur.ai/free/en/papers/wen-2025-stop-looking-important-tokens-multimodal/) - [next → Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoning](https://md.research.elamur.ai/free/en/papers/yang-2025-towards-thinking-optimal-scaling-test/) - [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/3/) ## References 32 references, 25 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. DeepSeek-AI Team. [DeepSeek-R1: Incentivizing reasoning capability in llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/), 2025. URL https://arxiv.org/abs/2501.12948. [↗](https://arxiv.org/abs/2501.12948) 2. 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](https://md.research.elamur.ai/free/en/papers/zeng-2025-simplerl-zoo-investigating-taming-zero/), 2025. URL https://arxiv.org/abs/2503.18892. [↗](https://arxiv.org/abs/2503.18892) 3. Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Tiantian Fan, Gaohong Liu, Lingjun Liu, Xin Liu, Haibin Lin, Zhiqi Lin, Bole Ma, Guangming Sheng, Yuxuan Tong, Chi Zhang, Mofan Zhang, Wang Zhang, Hang Zhu, Jinhua Zhu, Jiaze Chen, Jiangjie Chen, Chengyi Wang, Hongli Yu, Weinan D… — [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://md.research.elamur.ai/free/en/papers/yu-2025-dapo-open-source-llm-reinforcement/) 4. 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. Notion Blog. [find](https://scholar.google.com/scholar?q=Michael+Luo%2C+Sijun+Tan%2C+Justin+Wong%2C+Xiaoxiang+Shi%2C+William+Y%2E+Tang%2C+Manan+Roongta%2C+Colin+Cai%2C+Jeffrey+Luo%2C+Li+Erran+Li%2C+Raluca+Ada+Popa%2C+and+Ion+Stoica%2E+Deepscaler%3A+Surpassing+o1%2Dpreview+with+a+1%2E5b+) 5. Michael Luo, Sijun Tan, Roy Huang, Ameen Patel, Alpay Ariyak, Qingyang Wu, Xiaoxiang Shi, Rachel Xin, Colin Cai, Maurice Weber, Ce Zhang, Li Erran Li, Raluca Ada Popa, and Ion Stoica. Deepcoder: A fully open-source 14b coder at o3-mini level, 2025. Notion Blog. [find](https://scholar.google.com/scholar?q=Michael+Luo%2C+Sijun+Tan%2C+Roy+Huang%2C+Ameen+Patel%2C+Alpay+Ariyak%2C+Qingyang+Wu%2C+Xiaoxiang+Shi%2C+Rachel+Xin%2C+Colin+Cai%2C+Maurice+Weber%2C+Ce+Zhang%2C+Li+Erran+Li%2C+Raluca+Ada+Popa%2C+and+Ion+Stoica%2E+Deepcoder%3A+A+ful) 6. 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](https://md.research.elamur.ai/free/en/papers/shao-2024-deepseekmath-pushing-limits-mathematical-reasoning/), 2024. URL https://arxiv.org/abs/ 2402.03300. [↗](https://arxiv.org/abs/) 7. 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](https://md.research.elamur.ai/free/en/papers/muennighoff-2025-s1-simple-test-time-scaling/), 2025. URL https://arxiv.org/abs/2501.19393. [↗](https://arxiv.org/abs/2501.19393) 8. Yubo Wang, Xueguang Ma, Ge Zhang, Yuansheng Ni, Abhranil Chandra, Shiguang Guo, Weiming Ren, Aaran Arulraj, Xuan He, Ziyan Jiang, Tianle Li, Max Ku, Kai Wang, Alex Zhuang, Rongqi Fan, Xiang Yue, and Wenhu Chen. [MMLU-pro: A more robust and challenging multi-task language understanding benchmark](https://md.research.elamur.ai/free/en/papers/wang-2024-mmlu-pro-more-robust-challenging/). In T… 9. Xiang Yue, Tianyu Zheng, Ge Zhang, and Wenhu Chen. [MAmmoTH2: Scaling instructions from the web](https://md.research.elamur.ai/free/en/papers/yue-2024-mammoth2-scaling-instructions-web/). In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. URL [https://openreview.net/forum?id=yVu5dnPlqA](https://openreview.net/forum?id=yVu5dnPlqA). [↗](https://openreview.net/forum?id=yVu5dnPlqA) 10. 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](https://md.research.elamur.ai/free/en/papers/rein-2023-gpqa-graduate-level-google-proof/), 2023. URL https://arxiv.org/abs/2311.12022. [↗](https://arxiv.org/abs/2311.12022) 11. M-A-P. [SuperGPQA: Scaling LLM evaluation across 285 graduate disciplines](https://md.research.elamur.ai/free/en/papers/team-2025-supergpqa-scaling-llm-evaluation-across/), 2025. URL https://arxiv.org/abs/2502.14739. [↗](https://arxiv.org/abs/2502.14739) 12. 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](https://md.research.elamur.ai/free/en/papers/chen-2023-theoremqa-theorem-driven-question-answering/). In Houda Bouamor, Juan Pino, and Kalika Bali, editors, Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pa… 13. Mehran Kazemi, Bahare Fatemi, Hritik Bansal, John Palowitch, Chrysovalantis Anastasiou, Sanket Vaibhav Mehta, Lalit K. Jain, Virginia Aglietti, Disha Jindal, Peter Chen, Nishanth Dikkala, Gladys Tyen, Xin Liu, Uri Shalit, Silvia Chiappa, Kate Olszewska, Yi Tay, Vinh Q. Tran, Quoc V. Le, and Orhan Fi… — [BIG-Bench Extra Hard](https://md.research.elamur.ai/free/en/papers/kazemi-2025-big-bench-extra-hard/) 14. 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](https://md.research.elamur.ai/free/en/papers/hendrycks-2021-measuring-mathematical-problem-solving-math/). In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2), 2021. URL ht… 15. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, Christopher Hesse, and John Schulman. [Training verifiers to solve math word problems](https://md.research.elamur.ai/free/en/papers/cobbe-2021-training-verifiers-solve-math-word/), 2021. URL [https://arxiv.org/](https://arxiv.org/) abs/2110.14168. [↗](https://arxiv.org/) 16. 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](https://md.research.elamur.ai/free/en/papers/he-2024-olympiadbench-challenging-benchmark-promoting-agi/), 2024. URL… 17. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. [Chain-of-thought prompting elicits reasoning in large language models](https://md.research.elamur.ai/free/en/papers/wei-2022-chain-thought-prompting-elicits-reasoning/), 2023. URL https://arxiv.org/abs/2201.11903. [↗](https://arxiv.org/abs/2201.11903) 18. Qiyuan Zhang, Fuyuan Lyu, Zexu Sun, Lei Wang, Weixu Zhang, Wenyue Hua, Haolun Wu, Zhihan Guo, Yufei Wang, Niklas Muennighoff, Irwin King, Xue Liu, and Chen Ma. [A survey on test-time scaling in large language models: What, how, where, and how well?](https://md.research.elamur.ai/free/en/papers/zhang-2025-survey-test-time-scaling-large/), 2025. URL https://arxiv.org/abs/2503.24235. [↗](https://arxiv.org/abs/2503.24235) 19. Chengsong Huang, Langlin Huang, Jixuan Leng, Jiacheng Liu, and Jiaxin Huang. Efficient test-time scaling via self-calibration, 2025. URL https://arxiv.org/abs/2503.00031. — [arXiv:2503.00031](https://arxiv.org/abs/2503.00031) [↗](https://arxiv.org/abs/2503.00031) 20. OpenAI Team. [Openai o1 system card](https://md.research.elamur.ai/free/en/papers/openai-2024-openai-o1-system-card/), 2024. URL https://arxiv.org/abs/2412.16720. [↗](https://arxiv.org/abs/2412.16720) 21. Qwen Team. Qwen2.5 technical report, 2025. URL https://arxiv.org/abs/2412.15115. [↗](https://arxiv.org/abs/2412.15115) 22. Qwen Team. QwQ-32B: Embracing the power of reinforcement learning, March 2025. URL [https://qwenlm.github.io/blog/qwq-32b/](https://qwenlm.github.io/blog/qwq-32b/). [↗](https://qwenlm.github.io/blog/qwq-32b/) 23. Xiaomi LLM-Core Team. Mimo: Unlocking the reasoning potential of language model – from pretraining to posttraining. arXiv:2505.07608, 2025. URL https://arxiv.org/abs/2505. 07608. [↗](https://arxiv.org/abs/2505) 24. Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. Training la… — [Training language models to follow instructions with human feedback](https://md.research.elamur.ai/free/en/papers/ouyang-2022-training-language-models-follow-instructions/) 25. An Yang, Beichen Zhang, Binyuan Hui, Bofei Gao, Bowen Yu, Chengpeng Li, Dayiheng Liu, Jianhong Tu, Jingren Zhou, Junyang Lin, Keming Lu, Mingfeng Xue, Runji Lin, Tianyu Liu, Xingzhang Ren, and Zhenru Zhang. [Qwen2.5-math technical report: Toward mathematical expert model via self-improvement](https://md.research.elamur.ai/free/en/papers/yang-2024-qwen2-5-math-technical-report/), 2024. U… 26. Syeda Nahida Akter, Shrimai Prabhumoye, Matvei Novikov, Seungju Han, Ying Lin, Evelina Bakhturina, Eric Nyberg, Yejin Choi, Mostofa Patwary, Mohammad Shoeybi, and Bryan Catanzaro. [Nemotron-CrossThink: Scaling self-learning beyond math reasoning](https://md.research.elamur.ai/free/en/papers/akter-2026-nemotron-crossthink-scaling-self-learning/), 2025. URL https://arxiv.org/abs/2504.13941. [↗](https://arxiv.org/abs/2504.13941) 27. 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) [↗](https://arxiv.org/abs/2503.23829) 28. Gemini. [Gemini: A family of highly capable multimodal models](https://md.research.elamur.ai/free/en/papers/gemini-team-2023-gemini-family-highly-capable-multimodal/), 2024. URL https://arxiv.org/abs/2312.11805. [↗](https://arxiv.org/abs/2312.11805) 29. Mirac Suzgun, Nathan Scales, Nathanael Schärli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc V. Le, Ed H. Chi, Denny Zhou, and Jason Wei. [Challenging big-bench tasks and whether chain-of-thought can solve them](https://md.research.elamur.ai/free/en/papers/suzgun-2023-challenging-big-bench-tasks-whether/), 2022. URL https://arxiv.org/abs/2210.09261. [↗](https://arxiv.org/abs/2210.09261) 30. Aitor Lewkowycz, Anders Johan Andreassen, David Dohan, Ethan Dyer, Henryk Michalewski, Vinay Venkatesh Ramasesh, Ambrose Slone, Cem Anil, Imanol Schlag, Theo Gutman-Solo, Yuhuai Wu, Behnam Neyshabur, Guy Gur-Ari, and Vedant Misra. [Solving quantitative reasoning problems with language models](https://md.research.elamur.ai/free/en/papers/lewkowycz-2022-solving-quantitative-reasoning-problems-language/). In Alic… 31. OpenAI. [GPT-4o system card](https://md.research.elamur.ai/free/en/papers/openai-2024-gpt-4o-system-card/), 2024. URL https://arxiv.org/abs/2410.21276. [↗](https://arxiv.org/abs/2410.21276) 32. Jingcheng Hu, Yinmin Zhang, Qi Han, Daxin Jiang, Xiangyu Zhang, and Heung-Yeung Shum. [Open-reasoner-zero: An open source approach to scaling up reinforcement learning on the base model](https://md.research.elamur.ai/free/en/papers/hu-2025-open-reasoner-zero-open-source/), 2025. URL https://arxiv.org/abs/2503.24290. [↗](https://arxiv.org/abs/2503.24290) *Updated 2 November 2026*