---
title: "Tailored Primitive Initialization is the Secret Key to Reinforcement"
description: "A generator proposes a set of primitives. 79 references. Among them: 0."
authors: ["Yihang Yao", "Guangtao Zeng", "Raina Wu", "Yang Zhang", "Ding Zhao", "Zhang-Wei Hong", "Chuang Gan"]
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.1537/"
doi: "10.18653/v1/2026.acl-long.1537"
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/yao-2026-tailored-primitive-initialization-is-secret/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yao-2026-tailored-primitive-initialization-is-secret/)

[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/51/) / Tailored Primitive Initialization is the Secret Key to Reinforcement Learning

# Tailored Primitive Initialization is the Secret Key to Reinforcement Learning

**Authors:** Yihang Yao, [Guangtao Zeng](https://md.research.elamur.ai/free/en/authors/guangtao-zeng/), Raina Wu, [Yang Zhang](https://md.research.elamur.ai/free/en/authors/yang-zhang/), Ding Zhao, Zhang-Wei Hong and [Chuang Gan](https://md.research.elamur.ai/free/en/authors/chuang-gan/) · **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.1537/) · [PDF](https://aclanthology.org/2026.acl-long.1537.pdf) · [DOI 10.18653/v1/2026.acl-long.1537](https://doi.org/10.18653/v1/2026.acl-long.1537) · [OpenAlex](https://openalex.org/W4416355072)

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

## Summary

RL fine-tuning of LLMs for reasoning is sample-inefficient, and its gains depend heavily on the base model. Earlier work showed that a short SFT warm-up on traces with useful behaviours, such as reflection, helps. However, those behaviours were picked by hand for specific domains. The authors treat recurring patterns in thinking tokens as "reasoning primitives" (for example top-down or bottom-up solving). They argue that RL starts best from SFT data whose primitives are both high-quality and diverse. Quality means fitting the task and staying close to the student model's own distribution.

Their pipeline, Tailor, works in five steps:
1. The student model generates traces.
2. The traces are analysed.
3. A generator proposes a set of primitives.
4. A teacher uses the primitives as prompt instructions to build an SFT dataset.
5. The model is trained with RL.

On iGSM math tasks and the KK logic task, Tailor-initialized models reach higher accuracy after RL than rule-based, Standard-CoT, 4-STaR and BGKD baselines, including methods that rely on expert knowledge. This points to data-centric initialization as a practical way to make RL more efficient.

## Key points

- **Setup:** Base models include Llama3.2-1B and Qwen2.5-0.5B. RL uses DAPO by default with a binary rule-based outcome reward, and results are averaged over 3 random seeds.
- **Number of primitives:** The main runs use 25 primitives. Shrinking the set to 4, with total SFT size held constant, lowers post-RL accuracy on iGSM-medium. This indicates that primitive coverage matters more than wording diversity.
- **Diversity metric:** Tailor's SFT data has lower primitive-similarity scores than 4-STaR, meaning more diverse high-level reasoning. A high SFT pass@1 or high policy entropy alone did not predict good RL results.
- **Teacher temperature:** Raising the teacher's decoding temperature on KK generally helps, and Tailor beats the baselines at every setting tried. At t = 1.6, demonstration quality collapses and post-RL accuracy falls to near zero.
- **Other RL algorithms and limits:** The gains also hold with GRPO on iGSM-medium. The evaluation covers only verifiable, single-turn reasoning tasks.

*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

- [Massachusetts Institute of Technology](https://md.research.elamur.ai/free/en/universities/massachusetts-institute-of-technology/) — University, Cambridge, United States · 243 works · 76,506 citations
- [IBM](https://md.research.elamur.ai/free/en/companies/ibm/) — Company, Armonk, United States · 154 works · 72,606 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: Yihang Yao, Guangtao Zeng, Raina Wu, Yang Zhang, Ding Zhao, Zhang-Wei Hong and Chuang Gan. 2026. Tailored Primitive Initialization is the Secret Key to Reinforcement Learning. 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{yao2026tailored,
  title = {{Tailored Primitive Initialization is the Secret Key to Reinforcement Learning}},
  author = {Yihang Yao, Guangtao Zeng, Raina Wu, Yang Zhang, Ding Zhao, Zhang-Wei Hong and Chuang Gan},
  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.1537},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1537}
}
```

- [← previous TAGA@EEUCA 2026: Token-Attribution Guided Attention for Fine-Grained Toxic Behaviour Classification in Online Gaming Communities](https://md.research.elamur.ai/free/en/papers/shah-2026-taga-eeuca-2026-token-attribution/)
- [next → Tailoring Diagnostic Modeling to Individual Learners: Personalized Distractor Generation via MCTS-Guided Reasoning Reconstruction](https://md.research.elamur.ai/free/en/papers/wu-2026-tailoring-diagnostic-modeling-individual-learners/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/51/)

## References

79 references, 36 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Rishabh Agarwal, Nino Vieillard, Yongchao Zhou, Piotr Stanczyk, Sabela Ramos Garea, Matthieu Geist, and Olivier Bachem. 2024. [On-policy distillation of language models: Learning from self-generated mistakes](https://md.research.elamur.ai/free/en/papers/agarwal-2023-policy-distillation-language-models-learning/). In The twelfth international conference on learning representations.
2. Le, Sergey Levine, and Yi Ma. 2025. [Sft memorizes, rl generalizes: A comparative study of foundation model post-training](https://md.research.elamur.ai/free/en/papers/chu-2025-sft-memorizes-rl-generalizes-comparative/). arXiv preprint arXiv:2501.17161.
3. Simon S Du, Sham M Kakade, Ruosong Wang, and Lin F Yang. 2019. Is a good representation sufficient for sample efficient reinforcement learning? arXiv preprint arXiv:1910.03016. — [arXiv:1910.03016](https://arxiv.org/abs/1910.03016) [find](https://scholar.google.com/scholar?q=Simon+S+Du%2C+Sham+M+Kakade%2C+Ruosong+Wang%2C+and+Lin+F+Yang%2E+2019%2E+Is+a+good+representation+sufficient+for+sample+efficient+reinforcement+learning%3F+arXiv+preprint+arXiv%3A1910%2E03016%2E+%E2%80%94+%5BarXiv%3A1910%2E03016%5D%28ht)
4. Xidong Feng, Ziyu Wan, Muning Wen, Stephen Marcus McAleer, Ying Wen, Weinan Zhang, and Jun Wang. 2023. Alphazero-like tree-search can guide large language model decoding and training. arXiv preprint arXiv:2309.17179. [find](https://scholar.google.com/scholar?q=Xidong+Feng%2C+Ziyu+Wan%2C+Muning+Wen%2C+Stephen+Marcus+McAleer%2C+Ying+Wen%2C+Weinan+Zhang%2C+and+Jun+Wang%2E+2023%2E+Alphazero%2Dlike+tree%2Dsearch+can+guide+large+language+model+decoding+and+training%2E+arXiv+preprint+a)
5. Yuqian Fu, Tinghong Chen, Jiajun Chai, Xihuai Wang, Songjun Tu, Guojun Yin, Wei Lin, Qichao Zhang, Yuanheng Zhu, and Dongbin Zhao. 2025. Srft: A single-stage method with supervised and reinforcement fine-tuning for reasoning. arXiv preprint arXiv:2506.19767. — [arXiv:2506.19767](https://arxiv.org/abs/2506.19767) [find](https://scholar.google.com/scholar?q=Yuqian+Fu%2C+Tinghong+Chen%2C+Jiajun+Chai%2C+Xihuai+Wang%2C+Songjun+Tu%2C+Guojun+Yin%2C+Wei+Lin%2C+Qichao+Zhang%2C+Yuanheng+Zhu%2C+and+Dongbin+Zhao%2E+2025%2E+Srft%3A+A+single%2Dstage+method+with+supervised+and+reinforcement+f)
6. Kanishk Gandhi, Ayush Chakravarthy, Anikait Singh, Nathan Lile, and Noah D Goodman. 2025. [Cognitive behaviors that enable self-improving reasoners, or, four habits of highly effective stars](https://md.research.elamur.ai/free/en/papers/gandhi-2025-cognitive-behaviors-that-enable-self/). arXiv preprint arXiv:2503.01307.
7. Anirudh Goyal, Shagun Sodhani, Jonathan Binas, Xue Bin Peng, Sergey Levine, and Yoshua Bengio. 2020. Reinforcement learning with competitive ensembles of information-constrained primitives. In International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Anirudh+Goyal%2C+Shagun+Sodhani%2C+Jonathan+Binas%2C+Xue+Bin+Peng%2C+Sergey+Levine%2C+and+Yoshua+Bengio%2E+2020%2E+Reinforcement+learning+with+competitive+ensembles+of+information%2Dconstrained+primitives%2E+In+Interna)
8. Zhepeng Cen, Yihang Yao, William Han, Zuxin Liu, and Ding Zhao. 2025. Behavior injection: Preparing language models for reinforcement learning. arXiv preprint arXiv:2505.18917. — [arXiv:2505.18917](https://arxiv.org/abs/2505.18917) [find](https://scholar.google.com/scholar?q=Zhepeng+Cen%2C+Yihang+Yao%2C+William+Han%2C+Zuxin+Liu%2C+and+Ding+Zhao%2E+2025%2E+Behavior+injection%3A+Preparing+language+models+for+reinforcement+learning%2E+arXiv+preprint+arXiv%3A2505%2E18917%2E+%E2%80%94+%5BarXiv%3A2505%2E18917%5D%28ht)
9. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhishek Pandey, Ankur Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sravankumar, Artem Korenev, Arthur Hinsvark, and 11 othe… — [The Llama 3 Herd of Models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/)
10. Kevin Chen, Marco Cusumano-Towner, Brody Huval, Aleksei Petrenko, Jackson Hamburger, Vladlen Koltun, and Philipp Krähenbühl. 2025a. Reinforcement learning for long-horizon interactive llm agents. arXiv preprint arXiv:2502.01600. — [arXiv:2502.01600](https://arxiv.org/abs/2502.01600) [find](https://scholar.google.com/scholar?q=Kevin+Chen%2C+Marco+Cusumano%2DTowner%2C+Brody+Huval%2C+Aleksei+Petrenko%2C+Jackson+Hamburger%2C+Vladlen+Koltun%2C+and+Philipp+Kr%C3%A4henb%C3%BChl%2E+2025a%2E+Reinforcement+learning+for+long%2Dhorizon+interactive+llm+agents%2E+arXi)
11. Etash Guha, Ryan Marten, Sedrick Keh, Negin Raoof, Georgios Smyrnis, Hritik Bansal, Marianna Nezhurina, Jean Mercat, Trung Vu, Zayne Sprague, and 1 others. 2025. [Openthoughts: Data recipes for reasoning models](https://md.research.elamur.ai/free/en/papers/guha-2025-openthoughts-data-recipes-reasoning-models/). arXiv preprint arXiv:2506.04178.
12. Mingyang Chen, Tianpeng Li, Haoze Sun, Yijie Zhou, Chenzheng Zhu, Fan Yang, Zenan Zhou, Weipeng Chen, Haofen Wang, Jeff Z Pan, and 1 others. 2025b. [Learning to reason with search for llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/chen-2025-research-learning-reason-search-llms/). arXiv preprint arXiv:2503.19470.
13. Yutong Chen, Jiandong Gao, and Ji Wu. 2025c. Towards revealing the effectiveness of small-scale fine-tuning in r1-style reinforcement learning. arXiv preprint arXiv:2505.17988. — [arXiv:2505.17988](https://arxiv.org/abs/2505.17988) [find](https://scholar.google.com/scholar?q=Yutong+Chen%2C+Jiandong+Gao%2C+and+Ji+Wu%2E+2025c%2E+Towards+revealing+the+effectiveness+of+small%2Dscale+fine%2Dtuning+in+r1%2Dstyle+reinforcement+learning%2E+arXiv+preprint+arXiv%3A2505%2E17988%2E+%E2%80%94+%5BarXiv%3A2505%2E17988%5D%28ht)
14. Tianzhe Chu, Yuexiang Zhai, Jihan Yang, Shengbang Tong, Saining Xie, Dale Schuurmans, Quoc V Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, and 1 others. 2025. [Deepseek-r1: Incentivizing reasoning capability in llms via reinforc…](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/)
15. Tuomas Haarnoja, Aurick Zhou, Kristian Hartikainen, George Tucker, Sehoon Ha, Jie Tan, Vikash Kumar, Henry Zhu, Abhishek Gupta, Pieter Abbeel, and 1 others. 2018. Soft actor-critic algorithms and applications. arXiv preprint arXiv:1812.05905. [find](https://scholar.google.com/scholar?q=Tuomas+Haarnoja%2C+Aurick+Zhou%2C+Kristian+Hartikainen%2C+George+Tucker%2C+Sehoon+Ha%2C+Jie+Tan%2C+Vikash+Kumar%2C+Henry+Zhu%2C+Abhishek+Gupta%2C+Pieter+Abbeel%2C+and+1+others%2E+2018%2E+Soft+actor%2Dcritic+algorithms+and+appl)
16. Lixuan He, Jie Feng, and Yong Li. 2025. Amft: Aligning llm reasoners by meta-learning the optimal imitation-exploration balance. arXiv preprint arXiv:2508.06944. — [arXiv:2508.06944](https://arxiv.org/abs/2508.06944) [find](https://scholar.google.com/scholar?q=Lixuan+He%2C+Jie+Feng%2C+and+Yong+Li%2E+2025%2E+Amft%3A+Aligning+llm+reasoners+by+meta%2Dlearning+the+optimal+imitation%2Dexploration+balance%2E+arXiv+preprint+arXiv%3A2508%2E06944%2E+%E2%80%94+%5BarXiv%3A2508%2E06944%5D%28https%3A%2F%2Farxiv%2Eorg)
17. Zhang-Wei Hong, Pulkit Agrawal, Rémi Tachet des Combes, and Romain Laroche. 2023. Harnessing mixed offline reinforcement learning datasets via trajectory weighting. arXiv preprint arXiv:2306.13085. — [arXiv:2306.13085](https://arxiv.org/abs/2306.13085) [find](https://scholar.google.com/scholar?q=Zhang%2DWei+Hong%2C+Pulkit+Agrawal%2C+R%C3%A9mi+Tachet+des+Combes%2C+and+Romain+Laroche%2E+2023%2E+Harnessing+mixed+offline+reinforcement+learning+datasets+via+trajectory+weighting%2E+arXiv+preprint+arXiv%3A2306%2E13085%2E+%E2%80%94+)
18. Zhang-Wei Hong, Idan Shenfeld, Tsun-Hsuan Wang, Yung-Sung Chuang, Aldo Pareja, James Glass, Akash Srivastava, and Pulkit Agrawal. 2024. Curiosity-driven red-teaming for large language models. arXiv preprint arXiv:2402.19464. — [arXiv:2402.19464](https://arxiv.org/abs/2402.19464) [find](https://scholar.google.com/scholar?q=Zhang%2DWei+Hong%2C+Idan+Shenfeld%2C+Tsun%2DHsuan+Wang%2C+Yung%2DSung+Chuang%2C+Aldo+Pareja%2C+James+Glass%2C+Akash+Srivastava%2C+and+Pulkit+Agrawal%2E+2024%2E+Curiosity%2Ddriven+red%2Dteaming+for+large+language+models%2E+arXiv+pr)
19. Jian Hu. 2025. [Reinforce++: A simple and efficient approach for aligning large language models](https://md.research.elamur.ai/free/en/papers/hu-2025-reinforce-stabilizing-critic-free-policy/). arXiv preprint arXiv:2501.03262.
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. Bowen Jin, Hansi Zeng, Zhenrui Yue, Jinsung Yoon, Sercan Arik, Dong Wang, Hamed Zamani, and Jiawei Han. 2025. [Search-r1: Training llms to reason and leverage search engines with reinforcement learning](https://md.research.elamur.ai/free/en/papers/jin-2025-search-r1-training-llms-reason/). arXiv preprint arXiv:2503.09516.
22. Joongwon Kim, Anirudh Goyal, Liang Tan, Hannaneh Hajishirzi, Srinivasan Iyer, and Tianlu Wang. 2025. Astro: Teaching language models to reason by reflecting and backtracking in-context. arXiv preprint arXiv:2507.00417. — [arXiv:2507.00417](https://arxiv.org/abs/2507.00417) [find](https://scholar.google.com/scholar?q=Joongwon+Kim%2C+Anirudh+Goyal%2C+Liang+Tan%2C+Hannaneh+Hajishirzi%2C+Srinivasan+Iyer%2C+and+Tianlu+Wang%2E+2025%2E+Astro%3A+Teaching+language+models+to+reason+by+reflecting+and+backtracking+in%2Dcontext%2E+arXiv+preprint)
23. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph Gonzalez, Hao Zhang, and Ion Stoica. 2023. [Efficient memory management for large language model serving with pagedattention](https://md.research.elamur.ai/free/en/papers/kwon-2023-efficient-memory-management-large-language/). In Proceedings of the 29th symposium on operating systems principles, pages 611–626.
24. Chankyu Lee, Rajarshi Roy, Mengyao Xu, Jonathan Raiman, Mohammad Shoeybi, Bryan Catanzaro, and Wei Ping. 2024a. [Nv-embed: Improved techniques for training llms as generalist embedding models](https://md.research.elamur.ai/free/en/papers/lee-2024-nv-embed-improved-techniques-training/). arXiv preprint arXiv:2405.17428. Includes NV-Embed-v2; ranks No. 1 on the Massive Text Embedding Benchmark (…
25. Yang Li, Youssef Emad, Karthik Padthe, Jack Lanchantin, Weizhe Yuan, Thao Nguyen, Jason Weston, Shang-Wen Li, Dong Wang, Ilia Kulikov, and 1 others. 2025. Naturalthoughts: Selecting and distilling reasoning traces for general reasoning tasks. arXiv preprint arXiv:2507.01921. — [arXiv:2507.01921](https://arxiv.org/abs/2507.01921) [find](https://scholar.google.com/scholar?q=Yang+Li%2C+Youssef+Emad%2C+Karthik+Padthe%2C+Jack+Lanchantin%2C+Weizhe+Yuan%2C+Thao+Nguyen%2C+Jason+Weston%2C+Shang%2DWen+Li%2C+Dong+Wang%2C+Ilia+Kulikov%2C+and+1+others%2E+2025%2E+Naturalthoughts%3A+Selecting+and+distilling+rea)
26. Yiqing Liang, Jielin Qiu, Wenhao Ding, Zuxin Liu, James Tompkin, Mengdi Xu, Mengzhou Xia, Zhengzhong Tu, Laixi Shi, and Jiacheng Zhu. 2025. Modomodo: Multi-domain data mixtures for multimodal llm reinforcement learning. arXiv preprint arXiv:2505.24871. — [arXiv:2505.24871](https://arxiv.org/abs/2505.24871) [find](https://scholar.google.com/scholar?q=Yiqing+Liang%2C+Jielin+Qiu%2C+Wenhao+Ding%2C+Zuxin+Liu%2C+James+Tompkin%2C+Mengdi+Xu%2C+Mengzhou+Xia%2C+Zhengzhong+Tu%2C+Laixi+Shi%2C+and+Jiacheng+Zhu%2E+2025%2E+Modomodo%3A+Multi%2Ddomain+data+mixtures+for+multimodal+llm+rein)
27. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, Damai Dai, Daya Guo, Dejian Yang, Deli Chen, Dongjie Ji, Erhang Li, Fangyun Lin, Fucong Dai, Fuli Luo, and 159 others. 2024a. Deepseek-v3 technical report. [find](https://scholar.google.com/scholar?q=Aixin+Liu%2C+Bei+Feng%2C+Bing+Xue%2C+Bingxuan+Wang%2C+Bochao+Wu%2C+Chengda+Lu%2C+Chenggang+Zhao%2C+Chengqi+Deng%2C+Chenyu+Zhang%2C+Chong+Ruan%2C+Damai+Dai%2C+Daya+Guo%2C+Dejian+Yang%2C+Deli+Chen%2C+Dongjie+Ji%2C+Erhang+Li%2C+Fangyun)
28. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024b. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. [find](https://scholar.google.com/scholar?q=Aixin+Liu%2C+Bei+Feng%2C+Bing+Xue%2C+Bingxuan+Wang%2C+Bochao+Wu%2C+Chengda+Lu%2C+Chenggang+Zhao%2C+Chengqi+Deng%2C+Chenyu+Zhang%2C+Chong+Ruan%2C+and+1+others%2E+2024b%2E+Deepseek%2Dv3+technical+report%2E+arXiv+preprint+arXiv%3A241)
29. Mingjie Liu, Shizhe Diao, Ximing Lu, Jian Hu, Xin Dong, Yejin Choi, Jan Kautz, and Yi Dong. 2025. [Prorl: Prolonged reinforcement learning expands reasoning boundaries in large language models](https://md.research.elamur.ai/free/en/papers/liu-2025-prorl-prolonged-reinforcement-learning-expands/). arXiv preprint arXiv:2505.24864.
30. Zuxin Liu, Thai Hoang, Jianguo Zhang, Ming Zhu, Tian Lan, Juntao Tan, Weiran Yao, Zhiwei Liu, Yihao Feng, Rithesh RN, and 1 others. 2024c. Apigen: Automated pipeline for generating verifiable and diverse function-calling datasets. Advances in Neural Information Processing Systems, 37:54463–54482. [find](https://scholar.google.com/scholar?q=Zuxin+Liu%2C+Thai+Hoang%2C+Jianguo+Zhang%2C+Ming+Zhu%2C+Tian+Lan%2C+Juntao+Tan%2C+Weiran+Yao%2C+Zhiwei+Liu%2C+Yihao+Feng%2C+Rithesh+RN%2C+and+1+others%2E+2024c%2E+Apigen%3A+Automated+pipeline+for+generating+verifiable+and+dive)
31. Lu Ma, Hao Liang, Meiyi Qiang, Lexiang Tang, Xiaochen Ma, Zhen Hao Wong, Junbo Niu, Chengyu Shen, Runming He, Bin Cui, and 1 others. 2025. Learning what reinforcement learning can’t: Interleaved online fine-tuning for hardest questions. arXiv preprint arXiv:2506.07527. — [arXiv:2506.07527](https://arxiv.org/abs/2506.07527) [find](https://scholar.google.com/scholar?q=Lu+Ma%2C+Hao+Liang%2C+Meiyi+Qiang%2C+Lexiang+Tang%2C+Xiaochen+Ma%2C+Zhen+Hao+Wong%2C+Junbo+Niu%2C+Chengyu+Shen%2C+Runming+He%2C+Bin+Cui%2C+and+1+others%2E+2025%2E+Learning+what+reinforcement+learning+can%E2%80%99t%3A+Interleaved+onlin)
32. Subhabrata Mukherjee, Arindam Mitra, Ganesh Jawahar, Sahaj Agarwal, Hamid Palangi, and Ahmed Awadallah. 2023. [Orca: Progressive learning from complex explanation traces of gpt-4](https://md.research.elamur.ai/free/en/papers/mukherjee-2023-orca-progressive-learning-complex-explanation/). arXiv preprint arXiv:2306.02707.
33. Jaewoo Lee, Sujin Yun, Taeyoung Yun, and Jinkyoo Park. 2024b. Gta: Generative trajectory augmentation with guidance for offline reinforcement learning. Advances in Neural Information Processing Systems, 37:56766–56801. [find](https://scholar.google.com/scholar?q=Jaewoo+Lee%2C+Sujin+Yun%2C+Taeyoung+Yun%2C+and+Jinkyoo+Park%2E+2024b%2E+Gta%3A+Generative+trajectory+augmentation+with+guidance+for+offline+reinforcement+learning%2E+Advances+in+Neural+Information+Processing+System)
34. Xuan-Phi Nguyen, Shrey Pandit, Revanth Gangi Reddy, Austin Xu, Silvio Savarese, Caiming Xiong, and Shafiq Joty. 2025. Sfr-deepresearch: Towards effective reinforcement learning for autonomously reasoning single agents. arXiv preprint arXiv:2509.06283. — [arXiv:2509.06283](https://arxiv.org/abs/2509.06283) [find](https://scholar.google.com/scholar?q=Xuan%2DPhi+Nguyen%2C+Shrey+Pandit%2C+Revanth+Gangi+Reddy%2C+Austin+Xu%2C+Silvio+Savarese%2C+Caiming+Xiong%2C+and+Shafiq+Joty%2E+2025%2E+Sfr%2Ddeepresearch%3A+Towards+effective+reinforcement+learning+for+autonomously+reason)
35. Boyi Li, Yue Wang, Jiageng Mao, Boris Ivanovic, Sushant Veer, Karen Leung, and Marco Pavone. 2024. Driving everywhere with large language model policy adaptation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14948–14957. [find](https://scholar.google.com/scholar?q=Boyi+Li%2C+Yue+Wang%2C+Jiageng+Mao%2C+Boris+Ivanovic%2C+Sushant+Veer%2C+Karen+Leung%2C+and+Marco+Pavone%2E+2024%2E+Driving+everywhere+with+large+language+model+policy+adaptation%2E+In+Proceedings+of+the+IEEE%2FCVF+Confer)
36. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, and 1 others. 2022. [Training language models to follow instructions with human feedback](https://md.research.elamur.ai/free/en/papers/ouyang-2022-training-language-models-follow-instructions/). Advances in neural information processing systems, 35:27730–27744.
37. Jiayi Pan, Xingyao Wang, Graham Neubig, Navdeep Jaitly, Heng Ji, Alane Suhr, and Yizhe Zhang. 2024. [Training software engineering agents and verifiers with swe-gym](https://md.research.elamur.ai/free/en/papers/pan-2024-training-software-engineering-agents/). arXiv preprint arXiv:2412.21139.
38. Laixi Shi and Yuejie Chi. 2024. Distributionally robust model-based offline reinforcement learning with near-optimal sample complexity. Journal of Machine Learning Research, 25(200):1–91. [find](https://scholar.google.com/scholar?q=Laixi+Shi+and+Yuejie+Chi%2E+2024%2E+Distributionally+robust+model%2Dbased+offline+reinforcement+learning+with+near%2Doptimal+sample+complexity%2E+Journal+of+Machine+Learning+Research%2C+25%28200%29%3A1%E2%80%9391%2E)
39. Xue Bin Peng, Michael Chang, Grace Zhang, Pieter Abbeel, and Sergey Levine. 2019. Mcp: Learning composable hierarchical control with multiplicative compositional policies. Advances in neural information processing systems, 32. [find](https://scholar.google.com/scholar?q=Xue+Bin+Peng%2C+Michael+Chang%2C+Grace+Zhang%2C+Pieter+Abbeel%2C+and+Sergey+Levine%2E+2019%2E+Mcp%3A+Learning+composable+hierarchical+control+with+multiplicative+compositional+policies%2E+Advances+in+neural+informati)
40. Huatong Song, Jinhao Jiang, Yingqian Min, Jie Chen, Zhipeng Chen, Wayne Xin Zhao, Lei Fang, and Ji-Rong Wen. 2025. [R1-searcher: Incentivizing the search capability in llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/song-2025-r1-searcher-incentivizing-search/). arXiv preprint arXiv:2503.05592.
41. Akshara Prabhakar, Zuxin Liu, Ming Zhu, Jianguo Zhang, Tulika Awalgaonkar, Shiyu Wang, Zhiwei Liu, Haolin Chen, Thai Hoang, Juan Carlos Niebles, and 1 others. 2025. [Apigen-mt: Agentic pipeline for multi-turn data generation via simulated agent-human interplay](https://md.research.elamur.ai/free/en/papers/prabhakar-2025-apigen-mt-agentic-pipeline-multi/). arXiv preprint arXiv:2504.03601.
42. Yifan Sun, Jingyan Shen, Yibin Wang, Tianyu Chen, Zhendong Wang, Mingyuan Zhou, and Huan Zhang. 2025. Improving data efficiency for llm reinforcement fine-tuning through difficulty-targeted online data selection and rollout replay. arXiv preprint arXiv:2506.05316. — [arXiv:2506.05316](https://arxiv.org/abs/2506.05316) [find](https://scholar.google.com/scholar?q=Yifan+Sun%2C+Jingyan+Shen%2C+Yibin+Wang%2C+Tianyu+Chen%2C+Zhendong+Wang%2C+Mingyuan+Zhou%2C+and+Huan+Zhang%2E+2025%2E+Improving+data+efficiency+for+llm+reinforcement+fine%2Dtuning+through+difficulty%2Dtargeted+online+dat)
43. Martin L. Puterman. 2014. Markov Decision Processes: Discrete Stochastic Dynamic Programming, 2nd edition. John Wiley & Sons. [find](https://scholar.google.com/scholar?q=Martin+L%2E+Puterman%2E+2014%2E+Markov+Decision+Processes%3A+Discrete+Stochastic+Dynamic+Programming%2C+2nd+edition%2E+John+Wiley+%26+Sons%2E)
44. Zehan Qi, Xiao Liu, Iat Long Iong, Hanyu Lai, Xueqiao Sun, Wenyi Zhao, Yu Yang, Xinyue Yang, Jiadai Sun, Shuntian Yao, and 1 others. 2024. Webrl: Training llm web agents via self-evolving online curriculum reinforcement learning. arXiv preprint arXiv:2411.02337. — [arXiv:2411.02337](https://arxiv.org/abs/2411.02337) [find](https://scholar.google.com/scholar?q=Zehan+Qi%2C+Xiao+Liu%2C+Iat+Long+Iong%2C+Hanyu+Lai%2C+Xueqiao+Sun%2C+Wenyi+Zhao%2C+Yu+Yang%2C+Xinyue+Yang%2C+Jiadai+Sun%2C+Shuntian+Yao%2C+and+1+others%2E+2024%2E+Webrl%3A+Training+llm+web+agents+via+self%2Devolving+online+curri)
45. Cheng Qian, Emre Can Acikgoz, Qi He, Hongru Wang, Xiusi Chen, Dilek Hakkani-Tür, Gokhan Tur, and Heng Ji. 2025. [Toolrl: Reward is all tool learning needs](https://md.research.elamur.ai/free/en/papers/qian-2025-toolrl-reward-all-tool-learning/). arXiv preprint arXiv:2504.13958.
46. Yuxiao Qu, Anikait Singh, Yoonho Lee, Amrith Setlur, Ruslan Salakhutdinov, Chelsea Finn, and Aviral Kumar. 2025. Learning to discover abstractions for llm reasoning. In Proceedings of the AI4Math Workshop, ICML 2025. Poster. [find](https://scholar.google.com/scholar?q=Yuxiao+Qu%2C+Anikait+Singh%2C+Yoonho+Lee%2C+Amrith+Setlur%2C+Ruslan+Salakhutdinov%2C+Chelsea+Finn%2C+and+Aviral+Kumar%2E+2025%2E+Learning+to+discover+abstractions+for+llm+reasoning%2E+In+Proceedings+of+the+AI4Math+Work)
47. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, YK Li, Y 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.
48. Gemini Robotics Team, Saminda Abeyruwan, Joshua Ainslie, Jean-Baptiste Alayrac, Montserrat Gonzalez Arenas, Travis Armstrong, Ashwin Balakrishna, Robert Baruch, Maria Bauza, Michiel Blokzijl, and 1 others. 2025a. Gemini robotics: Bringing ai into the physical world. arXiv preprint arXiv:2503.20020. [find](https://scholar.google.com/scholar?q=Gemini+Robotics+Team%2C+Saminda+Abeyruwan%2C+Joshua+Ainslie%2C+Jean%2DBaptiste+Alayrac%2C+Montserrat+Gonzalez+Arenas%2C+Travis+Armstrong%2C+Ashwin+Balakrishna%2C+Robert+Baruch%2C+Maria+Bauza%2C+Michiel+Blokzijl%2C+and+1+ot)
49. Kimi Team, Angang Du, Bofei Gao, Bowei Xing, Changjiu Jiang, Cheng Chen, Cheng Li, Chenjun Xiao, Chenzhuang Du, Chonghua Liao, and 1 others. 2025b. 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/)
50. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 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/). Advances in neural information processing systems, 35:24824– 24837.
51. Yuxi Wei, Zi Wang, Yifan Lu, Chenxin Xu, Changxing Liu, Hao Zhao, Siheng Chen, and Yanfeng Wang. 2024. Editable scene simulation for autonomous driving via collaborative llm-agents. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 15077–15087. [find](https://scholar.google.com/scholar?q=Yuxi+Wei%2C+Zi+Wang%2C+Yifan+Lu%2C+Chenxin+Xu%2C+Changxing+Liu%2C+Hao+Zhao%2C+Siheng+Chen%2C+and+Yanfeng+Wang%2E+2024%2E+Editable+scene+simulation+for+autonomous+driving+via+collaborative+llm%2Dagents%2E+In+Proceedings+of+)
52. Haozhan Shen, Peng Liu, Jingcheng Li, Chunxin Fang, Yibo Ma, Jiajia Liao, Qiaoli Shen, Zilun Zhang, Kangjia Zhao, Qianqian Zhang, and 1 others. 2025a. [Vlm-r1: A stable and generalizable r1-style large vision-language model](https://md.research.elamur.ai/free/en/papers/shen-2025-vlm-r1-stable-generalizable-r1/). arXiv preprint arXiv:2504.07615.
53. Yuxiang Wei, Olivier Duchenne, Jade Copet, Quentin Carbonneaux, Lingming Zhang, Daniel Fried, Gabriel Synnaeve, Rishabh Singh, and Sida I Wang. 2025. [Swe-rl: Advancing llm reasoning via reinforcement learning on open software evolution](https://md.research.elamur.ai/free/en/papers/wei-2025-swe-rl-advancing-llm-reasoning/). arXiv preprint arXiv:2502.18449.
54. Maohao Shen, Guangtao Zeng, Zhenting Qi, Zhang-Wei Hong, Zhenfang Chen, Wei Lu, Gregory Wornell, Subhro Das, David Cox, and Chuang Gan. 2025b. Satori: Reinforcement learning with chain-of-action-thought enhances llm reasoning via autoregressive search. arXiv preprint arXiv:2502.02508. [find](https://scholar.google.com/scholar?q=Maohao+Shen%2C+Guangtao+Zeng%2C+Zhenting+Qi%2C+Zhang%2DWei+Hong%2C+Zhenfang+Chen%2C+Wei+Lu%2C+Gregory+Wornell%2C+Subhro+Das%2C+David+Cox%2C+and+Chuang+Gan%2E+2025b%2E+Satori%3A+Reinforcement+learning+with+chain%2Dof%2Daction%2Dthoug)
55. Xumeng Wen, Zihan Liu, Shun Zheng, Zhijian Xu, Shengyu Ye, Zhirong Wu, Xiao Liang, Yang Wang, Junjie Li, Ziming Miao, and 1 others. 2025. Reinforcement learning with verifiable rewards implicitly incentivizes correct reasoning in base llms. arXiv preprint arXiv:2506.14245. [find](https://scholar.google.com/scholar?q=Xumeng+Wen%2C+Zihan+Liu%2C+Shun+Zheng%2C+Zhijian+Xu%2C+Shengyu+Ye%2C+Zhirong+Wu%2C+Xiao+Liang%2C+Yang+Wang%2C+Junjie+Li%2C+Ziming+Miao%2C+and+1+others%2E+2025%2E+Reinforcement+learning+with+verifiable+rewards+implicitly+ince)
56. 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, pages 1279–1297.
57. Mingqi Wu, Zhihao Zhang, Qiaole Dong, Zhiheng Xi, Jun Zhao, Senjie Jin, Xiaoran Fan, Yuhao Zhou, Huijie Lv, Ming Zhang, and 1 others. 2025. Reasoning or memorization? unreliable results of reinforcement learning due to data contamination. arXiv preprint arXiv:2507.10532. — [arXiv:2507.10532](https://arxiv.org/abs/2507.10532) [find](https://scholar.google.com/scholar?q=Mingqi+Wu%2C+Zhihao+Zhang%2C+Qiaole+Dong%2C+Zhiheng+Xi%2C+Jun+Zhao%2C+Senjie+Jin%2C+Xiaoran+Fan%2C+Yuhao+Zhou%2C+Huijie+Lv%2C+Ming+Zhang%2C+and+1+others%2E+2025%2E+Reasoning+or+memorization%3F+unreliable+results+of+reinforceme)
58. Chunqiu Steven Xia, Yinlin Deng, Soren Dunn, and Lingming Zhang. 2024. [Agentless: Demystifying llm-based software engineering agents](https://md.research.elamur.ai/free/en/papers/xia-2024-agentless-demystifying-llm-based/). arXiv preprint arXiv:2407.01489.
59. Jason Weston, and Jing Xu. 2025a. Cot-self-instruct: Building high-quality synthetic prompts for reasoning and non-reasoning tasks. arXiv preprint arXiv:2507.23751. — [arXiv:2507.23751](https://arxiv.org/abs/2507.23751) [find](https://scholar.google.com/scholar?q=Jason+Weston%2C+and+Jing+Xu%2E+2025a%2E+Cot%2Dself%2Dinstruct%3A+Building+high%2Dquality+synthetic+prompts+for+reasoning+and+non%2Dreasoning+tasks%2E+arXiv+preprint+arXiv%3A2507%2E23751%2E+%E2%80%94+%5BarXiv%3A2507%2E23751%5D%28https%3A%2F%2Farxiv%2E)
60. Chulin Xie, Yangsibo Huang, Chiyuan Zhang, Da Yu, Xinyun Chen, Bill Yuchen Lin, Bo Li, Badih Ghazi, and Ravi Kumar. 2024. [On memorization of large language models in logical reasoning](https://md.research.elamur.ai/free/en/papers/xie-2025-memorization-large-language-models-logical/). arXiv preprint arXiv:2410.23123.
61. Ping Yu, Weizhe Yuan, Olga Golovneva, Tianhao Wu, Sainbayar Sukhbaatar, Jason Weston, and Jing Xu. 2025b. Rip: Better models by survival of the fittest prompts. arXiv preprint arXiv:2501.18578. — [arXiv:2501.18578](https://arxiv.org/abs/2501.18578) [find](https://scholar.google.com/scholar?q=Ping+Yu%2C+Weizhe+Yuan%2C+Olga+Golovneva%2C+Tianhao+Wu%2C+Sainbayar+Sukhbaatar%2C+Jason+Weston%2C+and+Jing+Xu%2E+2025b%2E+Rip%3A+Better+models+by+survival+of+the+fittest+prompts%2E+arXiv+preprint+arXiv%3A2501%2E18578%2E+%E2%80%94+%5BarX)
62. Jianhao Yan, Yafu Li, Zican Hu, Zhi Wang, Ganqu Cui, Xiaoye Qu, Yu Cheng, and Yue Zhang. 2025. [Learning to reason under off-policy guidance](https://md.research.elamur.ai/free/en/papers/yan-2025-learning-reason-under-off-policy/). arXiv preprint arXiv:2504.14945.
63. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, Kai Dang, and 23 others. 2024a. Qwen2.5 technical report. arXiv preprint arX… [find](https://scholar.google.com/scholar?q=An+Yang%2C+Baosong+Yang%2C+Beichen+Zhang%2C+Binyuan+Hui%2C+Bo+Zheng%2C+Bowen+Yu%2C+Chengyuan+Li%2C+Dayiheng+Liu%2C+Fei+Huang%2C+Haoran+Wei%2C+Huan+Lin%2C+Jian+Yang%2C+Jianhong+Tu%2C+Jianwei+Zhang%2C+Jianxin+Yang%2C+Jiaxi+Yang%2C+Jin)
64. John Yang, Carlos E Jimenez, Alexander Wettig, Kilian Lieret, Shunyu Yao, Karthik Narasimhan, and Ofir Press. 2024b. [Swe-agent: Agent-computer interfaces enable automated software engineering](https://md.research.elamur.ai/free/en/papers/yang-2024-swe-agent-agent-computer-interfaces/). Advances in Neural Information Processing Systems, 37:50528– 50652.
65. Yihang Yao, Zhepeng Cen, Wenhao Ding, Haohong Lin, Shiqi Liu, Tingnan Zhang, Wenhao Yu, and Ding Zhao. 2024. Oasis: Conditional distribution shaping for offline safe reinforcement learning. Advances in Neural Information Processing Systems, 37:78451– 78478. [find](https://scholar.google.com/scholar?q=Yihang+Yao%2C+Zhepeng+Cen%2C+Wenhao+Ding%2C+Haohong+Lin%2C+Shiqi+Liu%2C+Tingnan+Zhang%2C+Wenhao+Yu%2C+and+Ding+Zhao%2E+2024%2E+Oasis%3A+Conditional+distribution+shaping+for+offline+safe+reinforcement+learning%2E+Advances+i)
66. Tian Ye, Zicheng Xu, Yuanzhi Li, and Zeyuan Allen-Zhu. 2024. [Physics of language models: Part 2.1, grade-school math and the hidden reasoning process](https://md.research.elamur.ai/free/en/papers/ye-2024-physics-language-models-part-2/). In The Thirteenth International Conference on Learning Representations.
67. Tian Ye, Zicheng Xu, Yuanzhi Li, and Zeyuan Allen-Zhu. 2025. Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems. In Proceedings of the 13th International Conference on Learning Representations, ICLR ’25. Full version available at https://ssrn. com/abstract… [find](https://scholar.google.com/scholar?q=Tian+Ye%2C+Zicheng+Xu%2C+Yuanzhi+Li%2C+and+Zeyuan+Allen%2DZhu%2E+2025%2E+Physics+of+Language+Models%3A+Part+2%2E2%2C+How+to+Learn+From+Mistakes+on+Grade%2DSchool+Math+Problems%2E+In+Proceedings+of+the+13th+International+Co)
68. Edward Yeo, Yuxuan Tong, Morry Niu, Graham Neubig, and Xiang Yue. 2025. [Demystifying long chain-of-thought reasoning in llms](https://md.research.elamur.ai/free/en/papers/yeo-2025-demystifying-long-chain-thought-reasoning/). arXiv preprint arXiv:2502.03373.
69. Longhui Yu, Weisen Jiang, Han Shi, Jincheng Yu, Zhengying Liu, Yu Zhang, James T Kwok, Zhenguo Li, Adrian Weller, and Weiyang Liu. 2023a. [Metamath: Bootstrap your own mathematical questions for large language models](https://md.research.elamur.ai/free/en/papers/yu-2023-metamath-bootstrap-your-own-mathematical/). arXiv preprint arXiv:2309.12284.
70. Ping Yu, Jack Lanchantin, Tianlu Wang, Weizhe Yuan, Olga Golovneva, Ilia Kulikov, Sainbayar Sukhbaatar, Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Weinan Dai, Tiantian Fan, Gaohong Liu, Lingjun Liu, and 1 others. 2025c. [Dapo: An open-source llm reinforcement learning syst…](https://md.research.elamur.ai/free/en/papers/yu-2025-dapo-open-source-llm-reinforcement/)
71. Wenhao Yu, Nimrod Gileadi, Chuyuan Fu, Sean Kirmani, Kuang-Huei Lee, Montse Gonzalez Arenas, Hao-Tien Lewis Chiang, Tom Erez, Leonard Hasenclever, Jan Humplik, and 1 others. 2023b. Language to rewards for robotic skill synthesis. arXiv preprint arXiv:2306.08647. — [arXiv:2306.08647](https://arxiv.org/abs/2306.08647) [find](https://scholar.google.com/scholar?q=Wenhao+Yu%2C+Nimrod+Gileadi%2C+Chuyuan+Fu%2C+Sean+Kirmani%2C+Kuang%2DHuei+Lee%2C+Montse+Gonzalez+Arenas%2C+Hao%2DTien+Lewis+Chiang%2C+Tom+Erez%2C+Leonard+Hasenclever%2C+Jan+Humplik%2C+and+1+others%2E+2023b%2E+Language+to+rewards)
72. Yang Yue, Zhiqi Chen, Rui Lu, Andrew Zhao, Zhaokai Wang, Shiji Song, and Gao Huang. 2025a. [Does reinforcement learning really incentivize reasoning capacity in llms beyond the base model?](https://md.research.elamur.ai/free/en/papers/yue-2025-reinforcement-learning-really-incentivize-reasoning/) arXiv preprint arXiv:2504.13837.
73. Yu Yue, Yufeng Yuan, Qiying Yu, Xiaochen Zuo, Ruofei Zhu, Wenyuan Xu, Jiaze Chen, Chengyi Wang, TianTian Fan, Zhengyin Du, and 1 others. 2025b. [Vapo: Efficient and reliable reinforcement learning for advanced reasoning tasks](https://md.research.elamur.ai/free/en/papers/yue-2025-vapo-efficient-reliable-reinforcement-learning/). arXiv preprint arXiv:2504.05118.
74. Guangtao Zeng, Maohao Shen, Delin Chen, Zhenting Qi, Subhro Das, Dan Gutfreund, David Cox, Gregory Wornell, Wei Lu, Zhang-Wei Hong, and 1 others. 2025a. Satori-swe: Evolutionary test-time scaling for sample-efficient software engineering. arXiv preprint arXiv:2505.23604. — [arXiv:2505.23604](https://arxiv.org/abs/2505.23604) [find](https://scholar.google.com/scholar?q=Guangtao+Zeng%2C+Maohao+Shen%2C+Delin+Chen%2C+Zhenting+Qi%2C+Subhro+Das%2C+Dan+Gutfreund%2C+David+Cox%2C+Gregory+Wornell%2C+Wei+Lu%2C+Zhang%2DWei+Hong%2C+and+1+others%2E+2025a%2E+Satori%2Dswe%3A+Evolutionary+test%2Dtime+scaling+for+)
75. Weihao Zeng, Yuzhen Huang, Qian Liu, Wei Liu, Keqing He, Zejun Ma, and Junxian He. 2025b. [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/). arXiv preprint arXiv:2503.18892.
76. Kexun Zhang, Weiran Yao, Zuxin Liu, Yihao Feng, Zhiwei Liu, Rithesh RN, Tian Lan, Lei Li, Renze Lou, Jiacheng Xu, and 1 others. 2025. Diversity empowers intelligence: Integrating expertise of software engineering agents. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Kexun+Zhang%2C+Weiran+Yao%2C+Zuxin+Liu%2C+Yihao+Feng%2C+Zhiwei+Liu%2C+Rithesh+RN%2C+Tian+Lan%2C+Lei+Li%2C+Renze+Lou%2C+Jiacheng+Xu%2C+and+1+others%2E+2025%2E+Diversity+empowers+intelligence%3A+Integrating+expertise+of+software)
77. Rosie Zhao, Alexandru Meterez, Sham Kakade, Cengiz Pehlevan, Samy Jelassi, and Eran Malach. 2025. Echo chamber: Rl post-training amplifies behaviors learned in pretraining. arXiv preprint arXiv:2504.07912. [find](https://scholar.google.com/scholar?q=Rosie+Zhao%2C+Alexandru+Meterez%2C+Sham+Kakade%2C+Cengiz+Pehlevan%2C+Samy+Jelassi%2C+and+Eran+Malach%2E+2025%2E+Echo+chamber%3A+Rl+post%2Dtraining+amplifies+behaviors+learned+in+pretraining%2E+arXiv+preprint+arXiv%3A2504%2E0)
78. Haizhong Zheng, Yang Zhou, Brian R Bartoldson, Bhavya Kailkhura, Fan Lai, Jiawei Zhao, and Beidi Chen. 2025. Act only when it pays: Efficient reinforcement learning for llm reasoning via selective rollouts. arXiv preprint arXiv:2506.02177. — [arXiv:2506.02177](https://arxiv.org/abs/2506.02177) [find](https://scholar.google.com/scholar?q=Haizhong+Zheng%2C+Yang+Zhou%2C+Brian+R+Bartoldson%2C+Bhavya+Kailkhura%2C+Fan+Lai%2C+Jiawei+Zhao%2C+and+Beidi+Chen%2E+2025%2E+Act+only+when+it+pays%3A+Efficient+reinforcement+learning+for+llm+reasoning+via+selective+rol)
79. Chunting Zhou, Pengfei Liu, Puxin Xu, Srinivasan Iyer, Jiao Sun, Yuning Mao, Xuezhe Ma, Avia Efrat, Ping Yu, Lili Yu, and 1 others. 2023. Lima: Less is more for alignment. Advances in Neural Information Processing Systems, 36:55006–55021. [find](https://scholar.google.com/scholar?q=Chunting+Zhou%2C+Pengfei+Liu%2C+Puxin+Xu%2C+Srinivasan+Iyer%2C+Jiao+Sun%2C+Yuning+Mao%2C+Xuezhe+Ma%2C+Avia+Efrat%2C+Ping+Yu%2C+Lili+Yu%2C+and+1+others%2E+2023%2E+Lima%3A+Less+is+more+for+alignment%2E+Advances+in+Neural+Informati)

*Updated 2 November 2026*