Elamur Research⌕
☰
home / papers / 2025 / Intern-S1-MO: Long-horizon Reasoning Agent for Olympiad?Level Mathematical Problem Solving

Intern-S1-MO: Long-horizon Reasoning Agent for Olympiad?Level Mathematical Problem Solving

Paper · 2025 · Yuzhe Gu, Songyang Gao, Zijian Wu et al.

Summary

Even strong Large Reasoning Models (LRMs) struggle on IMO-level problems. The exploration these problems need outgrows a single 64K–128K context window, and earlier workarounds were mostly prompt-only pipelines on proprietary models. The authors present Intern-S1-MO, an agent framework that splits problem solving into many rounds. In each round, a reasoner attempts the problem and may stop at partial progress. A summarizer then condenses the trajectory into lemmas, and a verifier checks them before they enter a lemma library that feeds later rounds. The final draft is refined in a verify-and-revise loop. The underlying LRM is trained with OREAL-H, an online RL method derived from OREAL. It adds continuous rewards from an outcome process verifier (OPV) and is adapted to a hierarchical MDP. The system spends about 512K tokens per problem and reaches state-of-the-art scores, including a gold-level result in the official CMO2025. The work suggests that a compact lemma memory, rather than ever-longer contexts, can push LRMs toward Olympiad-grade proof writing.

Key points

  • On the five non-geometry IMO2025 problems, Intern-S1-MO scores 26/35, which is comparable to many silver medalists on that subset. This is not an official IMO medal.
  • In the official CMO2025, human experts graded its answers at 102/126, well above the gold threshold of 78. On CNMO2025 (14 problems), it scores 232.4/260.
  • It reports 96.6% pass@1 on AIME2025 and 95% on HMMT2025, outperforming leading LRMs.
  • Lemma summaries compress reasoning traces roughly 64:1, which extends the effective 64K budget about 8×.
  • In the ablation study, building up from a single-round baseline to the full system raises CNMO2025 from 178.0 to 232.4, and each component adds consistent gains.

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

🎓Shanghai Jiao Tong UniversityShanghai · China149 works · 16,161 citations🎓Renmin University of ChinaBeijing · China114 works · 16,118 citations🎓Chinese University of Hong KongHong Kong · China126 works · 14,073 citations🔬Shanghai Artificial Intelligence LaboratoryShanghai · China117 works · 11,281 citations🏛️Consejo Superior de Investigaciones CientíficasMadrid · Spain2 works · 16 citations

Also on the byline: Institute of Mathematical Sciences.

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksMathematics · 177 worksApplied Mathematics · 46 works

🧭 Topics 1Research subjects and languages the work is about.Reasoning · 1,426 works

Cite

Cite as Yuzhe Gu, Songyang Gao, Zijian Wu, Lingkai Kong, Wenwei Zhang, Zhongrui Cai, Fan Zheng, Tianyou Ma, Junhao Shen, Haiteng Zhao, Duanyang Zhang, Huilun Zhang, Kuikun Liu, Chengqi Lyu, Yanhui Duan, Chiyu Chen, Ningsheng Ma, Jianfei Gao, Han Lyu, Dahua Lin and Kai Chen. 2025. Intern-S1-MO: Long-horizon Reasoning Agent for Olympiad?Level Mathematical Problem Solving. arXiv:2512.10739.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2512.10739 · arXiv 2512.10739

@misc{gu2025intern,
  title = {{Intern-S1-MO: Long-horizon Reasoning Agent for Olympiad?Level Mathematical Problem Solving}},
  author = {Yuzhe Gu, Songyang Gao, Zijian Wu, Lingkai Kong, Wenwei Zhang, Zhongrui Cai, Fan Zheng, Tianyou Ma, Junhao Shen, Haiteng Zhao, Duanyang Zhang, Huilun Zhang, Kuikun Liu, Chengqi Lyu, Yanhui Duan, Chiyu Chen, Ningsheng Ma, Jianfei Gao, Han Lyu, Dahua Lin and Kai Chen},
  year = {2025},
  doi = {10.48550/arxiv.2512.10739},
  eprint = {2512.10739},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2512.10739}
}

References

54 references, 21 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Pranjal Aggarwal and Sean Welleck. L1: Controlling how long a reasoning model thinks with reinforcement learning. ArXiv, abs/2503.04697, 2025. URL https://api.semanticscholar.org/CorpusID:276813519. ↗
  2. Lei Bai, Zhongrui Cai, Yuhang Cao, Maosong Cao, Weihan Cao, Chiyu Chen, Haojiong Chen, Kai Chen, Pengcheng Chen, Ying Chen, et al. Intern-s1: A scientific multimodal foundation model. arXiv preprint arXiv:2508.15763, 2025.
  3. Mislav Balunovi’c, Jasper Dekoninck, Ivo Petrov, Nikola Jovanovi’c, and Martin T. Vechev. Matharena: Evaluating llms on uncontaminated math competitions. ArXiv, abs/2505.23281, 2025. URL https://api.semanticscholar.org/CorpusID:278996037. ↗
  4. Luoxin Chen, Jinming Gu, Liankai Huang, Wenhao Huang, Zhicheng Jiang, Allan Jie, Xiaoran Jin, Xing Jin, Chenggang Li, Kaijing Ma, et al. Seed-prover: Deep and broad reasoning for automated theorem proving. arXiv preprint arXiv:2507.23726, 2025. find
  5. Mark Chen. Evaluating large language models trained on code. arXiv:2107.03374, 2021.
  6. Google DeepMind. Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities. ArXiv, abs/2507.06261, 2025a. URL https://api.semanticscholar.org/CorpusID:280151524. ↗
  7. Google DeepMind. Advanced version of gemini with deep think officially achieves gold-medal standard at the international mathematical olympiad, 2025b. find
  8. Zhibin Gou, Zhihong Shao, Yeyun Gong, Yelong Shen, Yujiu Yang, Minlie Huang, Nan Duan, and Weizhu Chen. Tora: A tool-integrated reasoning agent for mathematical problem solving. ArXiv, abs/2309.17452, 2023. URL https://api.semanticscholar.org/CorpusID: 263310365. ↗
  9. Yuzhe Gu, Ziwei Ji, Wenwei Zhang, Chengqi Lyu, Dahua Lin, and Kai Chen. ANAH-v2: Scaling analytical hallucination annotation of large language models. In Advances in Neural Information Processing Systems, volume 37, pp. 60012–60039, 2024. doi: 10.52202/079017-1916. URL https://proceedings.neurips.cc… — [doi:10.52202/079017-1916](https://doi.org/10.52202/079017-1916) ↗
  10. Yuzhe Gu, Wenwei Zhang, Chengqi Lyu, Dahua Lin, and Kai Chen. Mask-DPO: Generalizable fine-grained factuality alignment of LLMs. arXiv preprint arXiv:2503.02846, 2025. URL https://arxiv.org/abs/2503.02846. — [arXiv:2503.02846](https://arxiv.org/abs/2503.02846) ↗
  11. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, et al. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948, 2025.
  12. Harvard–MIT Mathematics Tournament. Hmmt february problem archive. https://www.hmmt.org/www/tournaments/testing. Accessed 9 May 2025. ↗
  13. Zhouqi Hua, Wenwei Zhang, Chengqi Lyu, Yuzhe Gu, Songyang Gao, Kuikun Liu, Dahua Lin, and Kai Chen. The imitation game: Turing Machine imitator is length generalizable reasoner. arXiv preprint arXiv:2507.13332, 2025. URL https://arxiv.org/abs/2507.13332. — [arXiv:2507.13332](https://arxiv.org/abs/2507.13332) ↗
  14. Kaixuan Huang, Jiacheng Guo, Zihao Li, Xiang Ji, Jiawei Ge, Wenzhe Li, Yingqing Guo, Tianle Cai, Hui Yuan, Runzhe Wang, Yue Wu, Ming Yin, Shange Tang, Yangsibo Huang, Chi Jin, Xinyun Chen, Chiyuan Zhang, and Mengdi Wang. Math-perturb: Benchmarking llms’ math reasoning abilities against hard perturba… find
  15. Yichen Huang and Lin F Yang. Gemini 2.5 pro capable of winning gold at imo 2025. arXiv preprint arXiv:2507.15855, 2025. arXiv preprint Ziwei Ji, Yuzhe Gu, Wenwei Zhang, Chengqi Lyu, Dahua Lin, and Kai Chen. ANAH: Analytical annotation of hallucinations in large language models. In Proceedings of the… find
  16. Zixuan Ke, Fangkai Jiao, Yifei Ming, Xuan-Phi Nguyen, Austin Xu, Do Xuan Long, Minzhi Li, Chengwei Qin, PeiFeng Wang, Silvio Savarese, Caiming Xiong, and Shafiq Joty. A survey of frontiers in llm reasoning: Inference scaling, learning to reason, and agentic systems. Trans. Mach. Learn. Res., 2025, 2… find
  17. Jan Hendrik Kirchner, Yining Chen, Harri Edwards, Jan Leike, Nat McAleese, and Yuri Burda. Prover-verifier games improve legibility of llm outputs, 2024. URL https://arxiv.org/abs/2407.13692. — [arXiv:2407.13692](https://arxiv.org/abs/2407.13692) ↗
  18. Xiaoxi Li, Jiajie Jin, Guanting Dong, Hongjin Qian, Yutao Zhu, Yongkang Wu, Ji-Rong Wen, and Zhicheng Dou. Webthinker: Empowering large reasoning models with deep research capability. ArXiv, abs/2504.21776, 2025a. URL https://api.semanticscholar.org/CorpusID:278207550. ↗
  19. Xuefeng Li, Haoyang Zou, and Pengfei Liu. Torl: Scaling tool-integrated rl. arXiv preprint arXiv:2503.23383, 2025b. find
  20. Junlin Liu, Shengnan An, Shuang Zhou, Dan Ma, Yehao Lin, Xinxuan Lv, Xuanlin Wang, Xiaoyu Li, Ziwen Wang, Xuezhi Cao, and Xunliang Cai. AMO-bench: Large language models still struggle in high school math competitions. In Findings of the Association for Computational Linguistics: ACL 2026, pp. 2120–2… find
  21. Shudong Liu, Hongwei Liu, Junnan Liu, Linchen Xiao, Songyang Gao, Chengqi Lyu, Yuzhe Gu, Wenwei Zhang, Derek F Wong, Songyang Zhang, et al. Compassverifier: A unified and robust verifier for llms evaluation and outcome reward. In Proceedings of the 2025 Conference on Empirical Methods in Natural Lan…
  22. Ziyan Liu, Xueda Shen, Yuzhe Gu, Songyang Gao, Kuikun Liu, Guangran Cheng, Chengqi Lyu, Dahua Lin, Wenwei Zhang, and Kai Chen. ThoughtFold: Folding reasoning chains via introspective preference learning. arXiv preprint arXiv:2606.03503, 2026b. URL https://arxiv.org/abs/2606.03503. — [arXiv:2606.03503](https://arxiv.org/abs/2606.03503) ↗
  23. Chengqi Lyu, Songyang Gao, Yuzhe Gu, Wenwei Zhang, Jianfei Gao, Kuikun Liu, Ziyi Wang, Shuaibin Li, Qian Zhao, Haian Huang, et al. Exploring the limit of outcome reward for learning mathematical reasoning. arXiv preprint arXiv:2502.06781, 2025. — [arXiv:2502.06781](https://arxiv.org/abs/2502.06781) find
  24. Xinji Mai, Haotian Xu, Zhong-Zhi Li, Xing W, Weinong Wang, Jian Hu, Yingying Zhang, and Wenqiang Zhang. Agent rl scaling law: Agent rl with spontaneous code execution for mathematical problem solving, 2025. URL https://arxiv.org/abs/2505.07773. — [arXiv:2505.07773](https://arxiv.org/abs/2505.07773) ↗
  25. Mathematical Association of America. American invitational mathematics examination (aime) problems and solutions. https://maa.org/student-programs/amc/. Accessed 9 May 2025. ↗
  26. Sumeet Ramesh Motwani, Chandler Smith, Rocktim Jyoti Das, Markian Rybchuk, Philip Torr, Ivan Laptev, Fabio Pizzati, Ronald Clark, and Christian Schröder de Witt. Malt: Improving reasoning with multi-agent llm training. ArXiv, abs/2412.01928, 2024. URL https://api.semanticscholar.org/CorpusID:274446… — [arXiv:2412.01928](https://arxiv.org/abs/2412.01928) ↗
  27. 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. arXiv preprint arXiv:2501.19393, 2025.
  28. OpenAI. Introducing openai o3 and o4-mini, 2025a. URL https://openai.com/index/introducing-o3-and-o4-mini/. ↗
  29. OpenAI. gpt-oss-120b & gpt-oss-20b model card, 2025b. URL https://arxiv.org/abs/2508.10925. ↗
  30. OpenAI. Openai imo 2025 proofs, 2025c. openai-imo-2025-proofs. find
  31. Aske Plaat, Annie Wong, Suzan Verberne, Joost Broekens, Niki van Stein, and Thomas H.W. Back. Multi-step reasoning with large language models, a survey. 2024. URL — Large Language Models: A Survey https://api.semanticscholar.org/CorpusID:271218853. ↗
  32. Z. Z. Ren, Zhihong Shao, Jun-Mei Song, Huajian Xin, Haocheng Wang, Wanjia Zhao, Liyue Zhang, Zhe Fu, Qihao Zhu, Dejian Yang, Z. F. Wu, Zhibin Gou, Shirong Ma, Hongxuan Tang, Yuxuan Liu, Wenjun Gao, Daya Guo, and Chong Ruan. Deepseek-prover-v2: Advancing formal mathematical reasoning via reinforcemen… — DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition
  33. Ning Shang, Yifei Liu, Yi Zhu, Li Lyna Zhang, Weijiang Xu, Xinyu Guan, Buze Zhang, Bingcheng Dong, Xudong Zhou, Bowen Zhang, Ying Xin, Ziming Miao, Scarlett Li, Fan Yang, and Mao Yang. rstar2-agent: Agentic reasoning technical report, 2025. URL https://arxiv.org/abs/2508.20722. ↗
  34. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Jun-Mei Song, Mingchuan Zhang, Y. K. Li, Yu Wu, and Daya Guo. Deepseekmath: Pushing the limits of mathematical reasoning in open language models. ArXiv, abs/2402.03300, 2024. URL https://api.semanticscholar.org/CorpusID:267412607. ↗
  35. Chuming Shen, Wei Wei, Xiaoye Qu, and Yu Cheng. Satori-r1: Incentivizing multimodal reasoning with spatial grounding and verifiable rewards. arXiv preprint arXiv:2505.19094, 2025a. — [arXiv:2505.19094](https://arxiv.org/abs/2505.19094) find
  36. Junhao Shen, Haiteng Zhao, Yuzhe Gu, Songyang Gao, Kuikun Liu, Haian Huang, Jianfei Gao, Dahua Lin, Wenwei Zhang, and Kai Chen. Semi-off-policy reinforcement learning for vision-language slow-thinking reasoning. In Advances in Neural Information Processing Systems, volume 38, pp. 125082– 125115, 202… find
  37. Jiankai Sun, Chuanyang Zheng, Enze Xie, Zhengying Liu, Ruihang Chu, Jianing Qiu, Jiaqi Xu, Mingyu Ding, Hongyang Li, Mengzhe Geng, et al. A survey of reasoning with foundation models: Concepts, methodologies, and outlook. ACM Computing Surveys, 57 (11):1–43, 2025. find
  38. Lei Wang, Wanyu Xu, Yihuai Lan, Zhiqiang Hu, Yunshi Lan, Roy Ka-Wei Lee, and Ee-Peng Lim. Plan-and-solve prompting: Improving zero-shot chain-of-thought reasoning by large language models. In Annual Meeting of the Association for Computational Linguistics, 2023. URL https://api.semanticscholar.org/C… ↗
  39. Pengyuan Wang, Tian-Shuo Liu, Chenyang Wang, Yidi Wang, Shu Yan, Cheng-Xing Jia, Xu-Hui Liu, Xin-Wei Chen, Jia-Cheng Xu, Ziniu Li, and Yang Yu. A survey on large language models for mathematical reasoning. ArXiv, abs/2506.08446, 2025. URL https://api.semanticscholar.org/CorpusID:279261286. ↗
  40. Zijian Wu, Lingkai Kong, Wenwei Zhang, Songyang Gao, Yuzhe Gu, Zhongrui Cai, Tianyou Ma, Yuhong Liu, Zhi Wang, Runyuan Ma, et al. Opv: Outcome-based process verifier for efficient long chain-of-thought verification. arXiv preprint arXiv:2512.10756, 2025. xAI. Grok 4, 2025. URL https://x.ai/news/grok… — [arXiv:2512.10756](https://arxiv.org/abs/2512.10756) ↗
  41. URL — Qwen3 Technical Report https://github.com/aw31/An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, et al. Qwen3 technical report. arXiv preprint arXiv:2505.09388, 2025. ↗
  42. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, and Karthik Narasimhan. Tree of thoughts: Deliberate problem solving with large language models. ArXiv, abs/2305.10601, 2023. URL https://api.semanticscholar.org/CorpusID: 258762525. ↗
  43. Yurun Yuan and Tengyang Xie. Reinforce llm reasoning through multi-agent reflection. ArXiv, abs/2506.08379, 2025. URL https://api.semanticscholar.org/CorpusID: 279260670. — [arXiv:2506.08379](https://arxiv.org/abs/2506.08379) ↗
  44. Yang Yue, Zhiqi Chen, Rui Lu, Andrew Zhao, Zhaokai Wang, Shiji Song, and Gao Huang. Does reinforcement learning really incentivize reasoning capacity in llms beyond the base model? ArXiv, abs/2504.13837, 2025. URL https://api.semanticscholar.org/CorpusID: 277940134. ↗
  45. 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. ArXiv, abs/2503.18892, 2025. URL https://api.semanticscholar.org/CorpusID:277940848. ↗
  46. Dan Zhang, Sining Zhoubian, Yisong Yue, Yuxiao Dong, and Jie Tang. Rest-mcts*: Llm self-training via process reward guided tree search. ArXiv, abs/2406.03816, 2024a. URL https://api.semanticscholar.org/CorpusID:270285630. ↗
  47. Jianyi Zhang, Yufan Zhou, Jiuxiang Gu, Curtis Wigington, Tong Yu, Yiran Chen, Tong Sun, and Ruiyi Zhang. Artist: Improving the generation of text-rich images with disentangled diffusion models and large language models, 2024b. URL https://arxiv.org/abs/2406.12044. — [arXiv:2406.12044](https://arxiv.org/abs/2406.12044) ↗
  48. Zhuosheng Zhang, Aston Zhang, Mu Li, and Alexander J. Smola. Automatic chain of thought prompting in large language models. ArXiv, abs/2210.03493, 2022. URL https://api.semanticscholar.org/CorpusID:252762275. ↗
  49. Haiteng Zhao, Junhao Shen, Yiming Zhang, Songyang Gao, Kuikun Liu, Tianyou Ma, Fan Zheng, Dahua Lin, Wenwei Zhang, and Kai Chen. Achieving olympia-level geometry large language model agent via Complexity-Boosting Reinforcement Learning. arXiv preprint arXiv:2512.10534, 2025. URL https://arxiv.org/ab… — [arXiv:2512.10534](https://arxiv.org/abs/2512.10534) ↗
  50. Chujie Zheng, Zhenru Zhang, Beichen Zhang, Runji Lin, Keming Lu, Bowen Yu, Dayiheng Liu, Jingren Zhou, and Junyang Lin. ProcessBench: Identifying Process Errors in Mathematical Reasoning. arXiv e-prints, art. arXiv:2412.06559, December 2024. doi: 10.48550/arXiv.2412.06559.
  51. Denny Zhou, Nathanael Scharli, Le Hou, Jason Wei, Nathan Scales, Xuezhi Wang, Dale Schuurmans, Olivier Bousquet, Quoc Le, and Ed H. Chi. Least-to-most prompting enables complex reasoning in large language models. ArXiv, abs/2205.10625, 2022. URL https://api.semanticscholar.org/CorpusID:248986239. ↗
  52. Yichi Zhou, Jianqiu Zhao, Yongxin Zhang, Bohan Wang, Siran Wang, Luoxin Chen, Jiahui Wang, Haowei Chen, Allan Jie, Xinbo Zhang, Haocheng Wang, Luong Ngoc Trung, Rong Ye, Phan Nhat Hoang, Huishuai Zhang, Peng Sun, and Hang Li. Solving formal math problems by decomposition and iterative reflection. Ar… find
  53. Yicheng Zou, Dongsheng Zhu, Lin Zhu, Tong Zhu, Yunhua Zhou, Peiheng Zhou, Xinyu Zhou, Dongzhan Zhou, Zhiwang Zhou, Yuhao Zhou, et al. Intern-s1-pro: Scientific multimodal foundation model at trillion scale, 2026. URL https://arxiv.org/abs/2603.25040. ↗
  54. Yuxin Zuo, Kaiyan Zhang, Li Sheng, Shang Qu, Ganqu Cui, Xuekai Zhu, Haozhan Li, Yuchen Zhang, Xinwei Long, Ermo Hua, et al. Ttrl: Test-time reinforcement learning. arXiv preprint arXiv:2504.16084, 2025. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/gu-2025-intern-s1-mo-long/