---
title: "AdaMix: Adaptive Mixing for Short and Long Reasoning Adapters"
description: "Large Reasoning Models tend to produce long chain-of-thought traces even for easy questions. 0.7% accuracy. 46 references."
authors: ["Hao Luo", "Xiao Yan", "Xinyan Li", "Qiming Zeng", "Yuhao Lin", "Shanshan Feng", "Hao Wang", "Jiawei Jiang"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2026.acl-long.1864/"
doi: "10.18653/v1/2026.acl-long.1864"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/luo-2026-adamix-adaptive-mixing-short-long/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/luo-2026-adamix-adaptive-mixing-short-long/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/21/) / AdaMix: Adaptive Mixing for Short and Long Reasoning Adapters

# AdaMix: Adaptive Mixing for Short and Long Reasoning Adapters

**Authors:** [Hao Luo](https://md.research.elamur.ai/free/en/authors/hao-luo/), Xiao Yan, Xinyan Li, Qiming Zeng, Yuhao Lin, Shanshan Feng, [Hao Wang](https://md.research.elamur.ai/free/en/authors/hao-wang/) and Jiawei Jiang · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.acl-long.1864/) · [PDF](https://aclanthology.org/2026.acl-long.1864.pdf) · [DOI 10.18653/v1/2026.acl-long.1864](https://doi.org/10.18653/v1/2026.acl-long.1864) · [OpenAlex](https://openalex.org/W7166840940)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Large Reasoning Models tend to produce long chain-of-thought traces even for easy questions. This "overthinking" wastes compute and adds latency. Existing fixes have weaknesses. RL length penalties push one model toward two competing goals, model merging fixes the weights once for all inputs, and routing usually makes only a binary short-or-long choice. AdaMix addresses these issues by separating the two objectives. Training data are split into easy and hard subsets using LLM-based difficulty scoring. Two LoRA adapters are then trained independently with GRPO and dynamic rewards: a short adapter aimed at concise answers and a long adapter aimed at accurate deep reasoning. A BERT-based difficulty-aware router predicts a reasoning-intensity coefficient for each query, and this coefficient sets a task-arithmetic interpolation between the two adapters. On DeepSeek-R1-Distill-Qwen-7B across five math benchmarks, AdaMix cuts average response length by 54.9% and raises accuracy by up to 4.8%. The work shows that per-query weight mixing can offer finer control over reasoning cost than fixed merging or hard routing.

## Key points
- On GSM8K, responses become over 70% shorter with a 0.7% accuracy gain. On AIME25, length drops by about 42% while accuracy rises 4.8%.
- On an accuracy-efficiency score (AES), AdaMix outperforms competing methods like ShorterBetter and TLMRE with a margin of 30% or more.
- The BERT-based router (+1.2% accuracy, −54.2% tokens, AES 0.58) outperforms GPT-5 used as a router (−5.7% accuracy, −60.4% tokens, AES 0.50).
- Weight analysis shows the short adapter mainly changes shallow-to-middle layers and the long adapter mainly changes deeper ones. Their updates are close to orthogonal (cosine similarity near zero).
- The authors note two limitations. The router's sense of difficulty depends on the domain and needs recalibration for new domains, and the mixing weights come from a discrete set rather than a continuous range.

*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

- [Wuhan University](https://md.research.elamur.ai/free/en/universities/wuhan-university/) — University, Wuhan, China · 47 works · 1,246 citations

📚 Shelves (4): [🖥️ 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

🧭 Topics (1): [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,685 works

## Cite

Cite as: Hao Luo, Xiao Yan, Xinyan Li, Qiming Zeng, Yuhao Lin, Shanshan Feng, Hao Wang and Jiawei Jiang. 2026. AdaMix: Adaptive Mixing for Short and Long Reasoning Adapters. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{luo2026adamix,
  title = {{AdaMix: Adaptive Mixing for Short and Long Reasoning Adapters}},
  author = {Hao Luo, Xiao Yan, Xinyan Li, Qiming Zeng, Yuhao Lin, Shanshan Feng, Hao Wang and Jiawei Jiang},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1864},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1864}
}
```

- [← previous AdaJudge: Adaptive Multi-Perspective Judging for Reward Modeling](https://md.research.elamur.ai/free/en/papers/miao-2026-adajudge-adaptive-multi-perspective-judging/)
- [next → AdapShot: Adaptive Many-Shot In-Context Learning with Semantic-Aware KV Cache Reuse](https://md.research.elamur.ai/free/en/papers/ou-2026-adapshot-adaptive-many-shot-context/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/21/)

## References

46 references, 30 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. 2025. [L1: Controlling how long a reasoning model thinks with reinforcement learning](https://md.research.elamur.ai/free/en/papers/aggarwal-2025-l1-controlling-long-reasoning-model/). In Second Conference on Language Modeling.
2. Daman Arora and Andrea Zanette. 2025. [Training language models to reason efficiently](https://md.research.elamur.ai/free/en/papers/arora-2025-training-language-models-reason-efficiently/). In The Thirty-ninth Annual Conference on Neural Information Processing Systems.
3. Yupeng Chang, Xu Wang, Jindong Wang, Yuan Wu, Linyi Yang, Kaijie Zhu, Hao Chen, Xiaoyuan Yi, Cunxiang Wang, Yidong Wang, and 1 others. 2024. [A survey on evaluation of large language models](https://md.research.elamur.ai/free/en/papers/chang-2023-survey-evaluation-large-language-models/). ACM transactions on intelligent systems and technology.
4. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde De Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, and 1 others. 2021. [Evaluating large language models trained on code](https://md.research.elamur.ai/free/en/papers/chen-2021-evaluating-large-language-models-trained/). arXiv preprint arXiv:2107.03374.
5. Xingyu Chen, Jiahao Xu, Tian Liang, Zhiwei He, Jianhui Pang, Dian Yu, Linfeng Song, Qiuzhi Liu, Mengfei Zhou, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, and Dong Yu. 2025. Do not think that much for 2+3=? on the overthinking of long reasoning models. In Forty-second International Conference… [find](https://scholar.google.com/scholar?q=Xingyu+Chen%2C+Jiahao+Xu%2C+Tian+Liang%2C+Zhiwei+He%2C+Jianhui+Pang%2C+Dian+Yu%2C+Linfeng+Song%2C+Qiuzhi+Liu%2C+Mengfei+Zhou%2C+Zhuosheng+Zhang%2C+Rui+Wang%2C+Zhaopeng+Tu%2C+Haitao+Mi%2C+and+Dong+Yu%2E+2025%2E+Do+not+think+that+mu)
6. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, and 1 others. 2021. [Training verifiers to solve math word problems](https://md.research.elamur.ai/free/en/papers/cobbe-2021-training-verifiers-solve-math-word/). arXiv preprint arXiv:2110.14168.
7. Muzhi Dai, Shixuan Liu, and Qingyi Si. 2025. Stable reinforcement learning for efficient reasoning. arXiv preprint arXiv:2505.18086. — [arXiv:2505.18086](https://arxiv.org/abs/2505.18086) [find](https://scholar.google.com/scholar?q=Muzhi+Dai%2C+Shixuan+Liu%2C+and+Qingyi+Si%2E+2025%2E+Stable+reinforcement+learning+for+efficient+reasoning%2E+arXiv+preprint+arXiv%3A2505%2E18086%2E+%E2%80%94+%5BarXiv%3A2505%2E18086%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2505%2E18086%29)
8. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. [Bert: Pre-training of deep bidirectional transformers for language understanding](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/). In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies,…
9. Bofei Gao, Feifan Song, Zhe Yang, Zefan Cai, Yibo Miao, Qingxiu Dong, Lei Li, Chenghao Ma, Liang Chen, Runxin Xu, Zhengyang Tang, Benyou Wang, Daoguang Zan, Shanghaoran Quan, Ge Zhang, Lei Sha, Yichang Zhang, Xuancheng Ren, Tianyu Liu, and Baobao Chang. 2025. [Omni-math: A universal olympiad level ma…](https://md.research.elamur.ai/free/en/papers/gao-2024-omni-math-universal-olympiad-level/)
10. Chanhyuk Lee, Jiho Choi, Chanryeol Lee, Donggyun Kim, and Seunghoon Hong. 2025. Adarank: Adaptive rank pruning for enhanced model merging. arXiv preprint arXiv:2503.22178. — [arXiv:2503.22178](https://arxiv.org/abs/2503.22178) [find](https://scholar.google.com/scholar?q=Chanhyuk+Lee%2C+Jiho+Choi%2C+Chanryeol+Lee%2C+Donggyun+Kim%2C+and+Seunghoon+Hong%2E+2025%2E+Adarank%3A+Adaptive+rank+pruning+for+enhanced+model+merging%2E+arXiv+preprint+arXiv%3A2503%2E22178%2E+%E2%80%94+%5BarXiv%3A2503%2E22178%5D%28https%3A%2F)
11. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, and 1 others. 2025. Deepseekr1 incentivizes reasoning in llms through reinforcement learning. Nature. [find](https://scholar.google.com/scholar?q=Daya+Guo%2C+Dejian+Yang%2C+Haowei+Zhang%2C+Junxiao+Song%2C+Peiyi+Wang%2C+Qihao+Zhu%2C+Runxin+Xu%2C+Ruoyu+Zhang%2C+Shirong+Ma%2C+Xiao+Bi%2C+and+1+others%2E+2025%2E+Deepseekr1+incentivizes+reasoning+in+llms+through+reinforceme)
12. Hunter Lightman, Vineet Kosaraju, Yuri Burda, Harrison Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. 2024. [Let’s verify step by step](https://md.research.elamur.ai/free/en/papers/lightman-2023-let-s-verify-step-step/). In The Twelfth International Conference on Learning Representations.
13. Zhiwei He, Tian Liang, Jiahao Xu, Qiuzhi Liu, Xingyu Chen, Yue Wang, Linfeng Song, Dian Yu, Zhenwen Liang, Wenxuan Wang, and 1 others. 2025. [Deepmath-103k: A large-scale, challenging, decontaminated, and verifiable mathematical dataset for advancing reasoning](https://md.research.elamur.ai/free/en/papers/he-2025-deepmath-103k-large-scale-challenging/). arXiv preprint arXiv:2504.11456.
14. Yining Lu, Zilong Wang, Shiyang Li, Xin Liu, Changlong Yu, Qingyu Yin, Zhan Shi, Zixuan Zhang, and Meng Jiang. 2025. [Learning to optimize multi-objective alignment through dynamic reward weighting](https://md.research.elamur.ai/free/en/papers/lu-2026-learning-optimize-multi-objective-alignment/). arXiv preprint arXiv:2509.11452.
15. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021a. [Measuring massive multitask language understanding](https://md.research.elamur.ai/free/en/papers/hendrycks-2020-measuring-massive-multitask-language-understanding/). In International Conference on Learning Representations.
16. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. 2021b. [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).
17. Bairu Hou, Yang Zhang, Jiabao Ji, Yujian Liu, Kaizhi Qian, Jacob Andreas, and Shiyu Chang. 2025. [Thinkprune: Pruning long chain-of-thought of llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/hou-2025-thinkprune-pruning-long-chain-thought/). arXiv preprint arXiv:2504.01296.
18. Edward J Hu, yelong shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. [Lora: Low-rank adaptation of large language models](https://md.research.elamur.ai/free/en/papers/hu-2021-lora-low-rank-adaptation-large/). In International Conference on Learning Representations.
19. Gabriel Ilharco, Marco Tulio Ribeiro, Mitchell Wortsman, Ludwig Schmidt, Hannaneh Hajishirzi, and Ali Farhadi. 2023. [Editing models with task arithmetic](https://md.research.elamur.ai/free/en/papers/ilharco-2022-editing-models-task-arithmetic/). In The Eleventh International Conference on Learning Representations.
20. Aaron Jaech, Adam Kalai, Adam Lerer, Adam Richardson, Ahmed El-Kishky, Aiden Low, Alec Helyar, Aleksander Madry, Alex Beutel, Alex Carney, and 1 others. 2024. [Openai o1 system card](https://md.research.elamur.ai/free/en/papers/openai-2024-openai-o1-system-card/). arXiv preprint arXiv:2412.16720.
21. Naman Jain, King Han, Alex Gu, Wen-Ding Li, Fanjia Yan, Tianjun Zhang, Sida Wang, Armando Solar-Lezama, Koushik Sen, and Ion Stoica. 2025. [Livecodebench: Holistic and contamination free evaluation of large language models for code](https://md.research.elamur.ai/free/en/papers/jain-2024-livecodebench-holistic-contamination-free-evaluation/). In The Thirteenth International Conference on Learning Representatio…
22. Guosheng Liang, Longguang Zhong, Ziyi Yang, and Xiaojun Quan. 2025. Thinkswitcher: When to think hard, when to think fast. In Findings of the Association for Computational Linguistics: EMNLP 2025. [find](https://scholar.google.com/scholar?q=Guosheng+Liang%2C+Longguang+Zhong%2C+Ziyi+Yang%2C+and+Xiaojun+Quan%2E+2025%2E+Thinkswitcher%3A+When+to+think+hard%2C+when+to+think+fast%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+EMNLP+2025%2E)
23. Haotian Luo, Li Shen, Haiying He, Yibo Wang, Shiwei Liu, Wei Li, Naiqiang Tan, Xiaochun Cao, and Dacheng Tao. 2025. O1-pruner: Length-harmonizing fine-tuning for o1-like reasoning pruning. arXiv preprint arXiv:2501.12570. [find](https://scholar.google.com/scholar?q=Haotian+Luo%2C+Li+Shen%2C+Haiying+He%2C+Yibo+Wang%2C+Shiwei+Liu%2C+Wei+Li%2C+Naiqiang+Tan%2C+Xiaochun+Cao%2C+and+Dacheng+Tao%2E+2025%2E+O1%2Dpruner%3A+Length%2Dharmonizing+fine%2Dtuning+for+o1%2Dlike+reasoning+pruning%2E+arXiv+prepr)
24. Xinyin Ma, Guangnian Wan, Runpeng Yu, Gongfan Fang, and Xinchao Wang. 2025. [Cot-valve: Length-compressible chain-of-thought tuning](https://md.research.elamur.ai/free/en/papers/ma-2025-cot-valve-length-compressible-chain/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics.
25. Daniel Marczak, Simone Magistri, Sebastian Cygert, Bartłomiej Twardowski, Andrew D. Bagdanov, and Joost van de Weijer. 2025. No task left behind: Isotropic model merging with common and task-specific subspaces. In Forty-second International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=Daniel+Marczak%2C+Simone+Magistri%2C+Sebastian+Cygert%2C+Bart%C5%82omiej+Twardowski%2C+Andrew+D%2E+Bagdanov%2C+and+Joost+van+de+Weijer%2E+2025%2E+No+task+left+behind%3A+Isotropic+model+merging+with+common+and+task%2Dspecific+)
26. Isaac Ong, Amjad Almahairi, Vincent Wu, Wei-Lin Chiang, Tianhao Wu, Joseph E. Gonzalez, M Waleed Kadous, and Ion Stoica. 2025. Routellm: Learning to route llms from preference data. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Isaac+Ong%2C+Amjad+Almahairi%2C+Vincent+Wu%2C+Wei%2DLin+Chiang%2C+Tianhao+Wu%2C+Joseph+E%2E+Gonzalez%2C+M+Waleed+Kadous%2C+and+Ion+Stoica%2E+2025%2E+Routellm%3A+Learning+to+route+llms+from+preference+data%2E+In+The+Thirteenth+)
27. Haiquan Qiu, You Wu, Dong Li, Jianmin Guo, and Quanming Yao. 2025. Superpose task-specific features for model merging. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. [find](https://scholar.google.com/scholar?q=Haiquan+Qiu%2C+You+Wu%2C+Dong+Li%2C+Jianmin+Guo%2C+and+Quanming+Yao%2E+2025%2E+Superpose+task%2Dspecific+features+for+model+merging%2E+In+Proceedings+of+the+2025+Conference+on+Empirical+Methods+in+Natural+Language+Pr)
28. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. 2023. [Direct preference optimization: Your language model is secretly a reward model](https://md.research.elamur.ai/free/en/papers/rafailov-2023-direct-preference-optimization-language-model/). In Thirty-seventh Conference on Neural Information Processing Systems.
29. David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R. Bowman. 2024. [Gpqa: A graduate-level google-proof q&a benchmark](https://md.research.elamur.ai/free/en/papers/rein-2023-gpqa-graduate-level-google-proof/). In First Conference on Language Modeling.
30. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, YK Li, Yang Wu, and 1 others. 2024. [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/). arXiv preprint arXiv:2402.03300.
31. Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, and Chuan Wu. 2025. [Hybridflow: A flexible and efficient rlhf framework](https://md.research.elamur.ai/free/en/papers/sheng-2024-hybridflow-flexible-efficient-rlhf-framework/). In Proceedings of the Twentieth European Conference on Computer Systems.
32. Ying Sheng, Shiyi Cao, Dacheng Li, Coleman Hooper, Nicholas Lee, Shuo Yang, Christopher Chou, Banghua Zhu, Lianmin Zheng, Kurt Keutzer, Joseph E. Gonzalez, and Ion Stoica. 2024. Slora: Scalable serving of thousands of lora adapters. In Proceedings of Machine Learning and Systems. [find](https://scholar.google.com/scholar?q=Ying+Sheng%2C+Shiyi+Cao%2C+Dacheng+Li%2C+Coleman+Hooper%2C+Nicholas+Lee%2C+Shuo+Yang%2C+Christopher+Chou%2C+Banghua+Zhu%2C+Lianmin+Zheng%2C+Kurt+Keutzer%2C+Joseph+E%2E+Gonzalez%2C+and+Ion+Stoica%2E+2024%2E+Slora%3A+Scalable+servin)
33. George Stoica, Pratik Ramesh, Boglarka Ecsedi, Leshem Choshen, and Judy Hoffman. 2025. Model merging with svd to tie the knots. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=George+Stoica%2C+Pratik+Ramesh%2C+Boglarka+Ecsedi%2C+Leshem+Choshen%2C+and+Judy+Hoffman%2E+2025%2E+Model+merging+with+svd+to+tie+the+knots%2E+In+The+Thirteenth+International+Conference+on+Learning+Representations%2E)
34. Yang Sui, Yu-Neng Chuang, Guanchu Wang, Jiamu Zhang, Tianyi Zhang, Jiayi Yuan, Hongyi Liu, Andrew Wen, Shaochen Zhong, Na Zou, Hanjie Chen, and Xia Hu. 2025. [Stop overthinking: A survey on efficient reasoning for large language models](https://md.research.elamur.ai/free/en/papers/sui-2025-stop-overthinking-survey-efficient-reasoning/). Transactions on Machine Learning Research.
35. Kimi Team, Angang Du, Bofei Gao, Bowei Xing, Changjiu Jiang, Cheng Chen, Cheng Li, Chenjun Xiao, Chenzhuang Du, Chonghua Liao, and 1 others. 2025. Kimi k1. 5: Scaling reinforcement learning with llms. arXiv preprint arXiv:2501.12599. — [Kimi k1.5: Scaling Reinforcement Learning with LLMs](https://md.research.elamur.ai/free/en/papers/kimi-team-2025-kimi-k1-5-scaling-reinforcement/)
36. Chen Wang, Xunzhuo Liu, Yuhan Liu, Yue Zhu, Xiangxi Mo, Junchen Jiang, and Huamin Chen. 2025. When to reason: Semantic router for vllm. arXiv preprint arXiv:2510.08731. 2025a. Qwen3 technical report. arXiv preprint arXiv:2505.09388. — [arXiv:2510.08731](https://arxiv.org/abs/2510.08731) [find](https://scholar.google.com/scholar?q=Chen+Wang%2C+Xunzhuo+Liu%2C+Yuhan+Liu%2C+Yue+Zhu%2C+Xiangxi+Mo%2C+Junchen+Jiang%2C+and+Huamin+Chen%2E+2025%2E+When+to+reason%3A+Semantic+router+for+vllm%2E+arXiv+preprint+arXiv%3A2510%2E08731%2E+2025a%2E+Qwen3+technical+report%2E+)
37. An Yang, Beichen Zhang, Binyuan Hui, Bofei Gao, Bowen Yu, Chengpeng Li, Dayiheng Liu, Jianhong Tu, Jingren Zhou, Junyang Lin, and 1 others. 2024. [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/). arXiv preprint arXiv:2409.12122.
38. Wenkai Yang, Shuming Ma, Yankai Lin, and Furu Wei. 2025b. [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/). In The Thirty-ninth Annual Conference on Neural Information Processing Systems.
39. Jingyang Yi, Justin Wang, and Sida Li. 2025. Shorterbetter: Guiding reasoning models to find optimal inference length for efficient reasoning. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Jingyang+Yi%2C+Justin+Wang%2C+and+Sida+Li%2E+2025%2E+Shorterbetter%3A+Guiding+reasoning+models+to+find+optimal+inference+length+for+efficient+reasoning%2E+In+The+Thirty%2Dninth+Annual+Conference+on+Neural+Informati)
40. Le Yu, Bowen Yu, Haiyang Yu, Fei Huang, and Yongbin Li. 2024. [Language models are super mario: Absorbing abilities from homologous models as a free lunch](https://md.research.elamur.ai/free/en/papers/yu-2023-language-models-super-mario-absorbing/). In Forty-first International Conference on Machine Learning.
41. Duzhen Zhang, Zhong-Zhi Li, Ming-Liang Zhang, Jiaxin Zhang, Zengyan Liu, Yuxuan Yao, Haotian Xu, Junhao Zheng, Xiuyi Chen, Yingying Zhang, Fei Yin, Jiahua Dong, Zhijiang Guo, Le Song, and Cheng-Lin Liu. 2025a. [From system 1 to system 2: A survey of reasoning large language models](https://md.research.elamur.ai/free/en/papers/li-2025-system-1-system-2-survey/). IEEE Transactions…
42. Ruiqi Zhang, Changyi Xiao, and Yixin Cao. 2025b. Long or short cot? investigating instance-level switch of large reasoning models. arXiv preprint arXiv:2506.04182. — [arXiv:2506.04182](https://arxiv.org/abs/2506.04182) [find](https://scholar.google.com/scholar?q=Ruiqi+Zhang%2C+Changyi+Xiao%2C+and+Yixin+Cao%2E+2025b%2E+Long+or+short+cot%3F+investigating+instance%2Dlevel+switch+of+large+reasoning+models%2E+arXiv+preprint+arXiv%3A2506%2E04182%2E+%E2%80%94+%5BarXiv%3A2506%2E04182%5D%28https%3A%2F%2Farxiv%2Eo)
43. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, brian ichter, Fei Xia, Ed H. Chi, Quoc V Le, and Denny Zhou. 2022. [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/). In Advances in Neural Information Processing Systems.
44. Han Wu, Yuxuan Yao, Shuqi Liu, Zehua Liu, Xiaojin Fu, Xiongwei Han, Xing Li, Hui-Ling Zhen, Tao Zhong, and Mingxuan Yuan. 2025. Unlocking efficient longto-short llm reasoning with model merging. arXiv preprint arXiv:2503.20641. — [arXiv:2503.20641](https://arxiv.org/abs/2503.20641) [find](https://scholar.google.com/scholar?q=Han+Wu%2C+Yuxuan+Yao%2C+Shuqi+Liu%2C+Zehua+Liu%2C+Xiaojin+Fu%2C+Xiongwei+Han%2C+Xing+Li%2C+Hui%2DLing+Zhen%2C+Tao+Zhong%2C+and+Mingxuan+Yuan%2E+2025%2E+Unlocking+efficient+longto%2Dshort+llm+reasoning+with+model+merging%2E+arXiv)
45. Prateek Yadav, Derek Tam, Leshem Choshen, Colin Raffel, and Mohit Bansal. 2023. [Ties-merging: Resolving interference when merging models](https://md.research.elamur.ai/free/en/papers/yadav-2023-ties-merging-resolving-interference-merging/). In Thirty-seventh Conference on Neural Information Processing Systems.
46. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. A PPENDIX A Additional Experimental Details A.1 Environment . . . . . . . . . . . . A.2 Dataset Details . . . . . . . . . . A.2.1 Training Dataset. . . . . . A.2.2 Eva… [find](https://scholar.google.com/scholar?q=An+Yang%2C+Anfeng+Li%2C+Baosong+Yang%2C+Beichen+Zhang%2C+Binyuan+Hui%2C+Bo+Zheng%2C+Bowen+Yu%2C+Chang+Gao%2C+Chengen+Huang%2C+Chenxu+Lv%2C+and+1+others%2E+A+PPENDIX+A+Additional+Experimental+Details+A%2E1+Environment+%2E+%2E+%2E+%2E)

*Updated 2 November 2026*