---
title: "A Survey of On-Policy Distillation for Large Language Models"
description: "The authors do not introduce new data or a new model. 228 references, cited by 41 papers here."
authors: ["Mingyang Song", "Mao Zheng"]
year: 2026
venue: "arXiv:2604.00626"
source: "https://arxiv.org/abs/2604.00626"
doi: "10.48550/arxiv.2604.00626"
arxiv: "2604.00626"
licence: "arXiv — non-exclusive license to distribute"
licence_url: "http://arxiv.org/licenses/nonexclusive-distrib/1.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/song-2026-survey-policy-distillation-large-language/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/song-2026-survey-policy-distillation-large-language/)

[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/) / A Survey of On-Policy Distillation for Large Language Models

# A Survey of On-Policy Distillation for Large Language Models

**Authors:** [Mingyang Song](https://md.research.elamur.ai/free/en/authors/mingyang-song/) and [Mao Zheng](https://md.research.elamur.ai/free/en/authors/mao-zheng/) · **Year:** 2026 · **Venue:** arXiv:2604.00626 · **Licence:** [arXiv — non-exclusive license to distribute](http://arxiv.org/licenses/nonexclusive-distrib/1.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2604.00626) · [PDF](https://arxiv.org/pdf/2604.00626) · [DOI 10.48550/arxiv.2604.00626](https://doi.org/10.48550/arxiv.2604.00626) · [arXiv 2604.00626](https://arxiv.org/abs/2604.00626) · [OpenAlex](https://openalex.org/W7148454362)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [41](https://md.research.elamur.ai/free/en/papers/song-2026-survey-policy-distillation-large-language/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Distilling large LLMs into smaller ones usually means training the student on text the teacher already wrote. At inference, though, the student continues from its own imperfect tokens, and these early mistakes snowball over long outputs. This exposure bias hurts reasoning tasks most. The survey treats On-Policy Distillation (OPD) as its own paradigm. In OPD, the student samples its own sequences, and a teacher (or the model itself) scores or corrects them.

The authors do not introduce new data or a new model. Their contribution is a synthesis of more than two hundred papers, which they cast as f-divergence minimization over student-generated trajectories. They sort methods along three axes: training objective, where the feedback signal comes from (white-box, black-box, or teacher-free), and how training is stabilized. They also catalogue typical failure modes and show that OPD is formally linked to KL-constrained RL.

This matters because OPD is now used in production pipelines. For smaller-language projects, its teacher-free and multilingual safety variants are especially relevant.

## Key points
- The theoretical motivation comes from imitation learning (DAgger). Off-policy imitation accumulates error on the order of O(ϵT²) over horizon T. Querying the expert on states the learner actually visits reduces this to O(ϵT).
- GKD and MiniLLM launched LLM-era OPD in mid-2023. The field has since grown to over 200 papers, adding more than 10 per month through mid-2026. A companion list is maintained at the GitHub repo nick7nlp/Awesome-LLM-On-Policy-Distillation.
- Industrial adopters include Qwen3, Gemma 2, MiMo-V2-Flash and DeepSeek-V4. DeepSeek-V4 replaced its mixed RL stage entirely with multi-teacher OPD.
- The documented failure modes are diversity collapse, the trap posed by flawed prefixes, saturation during self-play, and a gap between calibration and capability. OPD costs roughly 4–5× the compute of off-policy SFT. Speculative decoding, prefix truncation or offline precomputation can reduce this overhead.
- Teacher-free self-distillation methods (OPSD, SD-ZERO) need no external teacher. MSD carries safety alignment from high-resource to low-resource languages without any target-language safety data.

*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

- [Tencent](https://md.research.elamur.ai/free/en/companies/tencent/) — Company, Shenzhen, China · 174 works · 21,355 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 19,503 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 18,805 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,959 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,789 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 4,145 works · [🪶 Efficient models & compression](https://md.research.elamur.ai/free/en/topics/efficient-models/) · 328 works

## Cite

Cite as: Mingyang Song and Mao Zheng. 2026. A Survey of On-Policy Distillation for Large Language Models. arXiv:2604.00626.

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

```bibtex
@misc{song2026survey,
  title = {{A Survey of On-Policy Distillation for Large Language Models}},
  author = {Mingyang Song and Mao Zheng},
  year = {2026},
  doi = {10.48550/arxiv.2604.00626},
  eprint = {2604.00626},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2604.00626}
}
```

- [← previous Entropy-Aware On-Policy Distillation of Language Models](https://md.research.elamur.ai/free/en/papers/jin-2026-entropy-aware-policy-distillation-language/)
- [next → SemEval-2026 Task 13: Detecting Machine-Generated Code with Multiple Programming Languages, Generators, and Application Scenarios](https://md.research.elamur.ai/free/en/papers/orel-2026-semeval-2026-task-13-detecting/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/)

## References

228 references, 49 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Amirhossein Afsharrad, Amirhesam Abedsoltan, Ahmadreza Moradipari, and Sanjay Lall. On-Policy Distillation of Language Models for Autonomous Vehicle Motion Planning. arXiv preprint arXiv:2604.07944, 2026. URL https://arxiv.org/abs/2604.07944. — [arXiv:2604.07944](https://arxiv.org/abs/2604.07944) [↗](https://arxiv.org/abs/2604.07944)
2. Rishabh Agarwal, Nino Vieillard, Yongchao Zhou, Piotr Stanczyk, Sabela Ramos, Matthieu Geist, and Olivier Bachem. [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/). Proceedings of ICLR, 2024. URL https://arxiv.org/abs/2306.13649. [↗](https://arxiv.org/abs/2306.13649)
3. Rishabh Agrawal, Jacob Fein-Ashley, and Paria Rashidinejad. Reinforcement Learning from Rich Feedback with Distributional DAgger. arXiv preprint arXiv:2606.05152, 2026. URL https://arxiv.org/abs/2606.05152. — [arXiv:2606.05152](https://arxiv.org/abs/2606.05152) [↗](https://arxiv.org/abs/2606.05152)
4. Mohammadreza Armandpour, Fatih Ilhan, David Harrison, Ajay Jaiswal, Duc N. M Hoang, Fartash Faghri, Yizhe Zhang, Minsik Cho, and Mehrdad Farajtabar. [Unmasking on-policy distillation: Where it helps, where it hurts, and why](https://md.research.elamur.ai/free/en/papers/armandpour-2026-unmasking-policy-distillation-where-helps/). arXiv preprint arXiv:2605.10889v1, 2026.
5. Nicolas Boizard, Kevin El Haddad, Céline Hudelot, and Pierre Colombo. [Towards Cross-Tokenizer Distillation: the Universal Logit Distillation Loss for LLMs](https://md.research.elamur.ai/free/en/papers/boizard-2024-towards-cross-tokenizer-distillation-universal/). Transactions on Machine Learning Research, 2025. URL https://arxiv.org/abs/2402.12030. [↗](https://arxiv.org/abs/2402.12030)
6. Walid Bousselham, Hilde Kuehne, and Cordelia Schmid. VOLD: Reasoning Transfer from LLMs to Vision-Language Models via On-Policy Distillation. arXiv preprint arXiv:2510.23497, 2025. URL https://arxiv.org/abs/2510.23497. — [arXiv:2510.23497](https://arxiv.org/abs/2510.23497) [↗](https://arxiv.org/abs/2510.23497)
7. Dan Busbridge, Amitis Shidani, Floris Weers, Jason Ramapuram, Etai Littwin, and Russ Webb. [Distillation Scaling Laws](https://md.research.elamur.ai/free/en/papers/busbridge-2025-distillation-scaling-laws/). In Proceedings of the International Conference on Machine Learning (ICML), 2025. URL https://arxiv.org/abs/2502.08606. [↗](https://arxiv.org/abs/2502.08606)
8. Woohyeon Byeon, Jiwon Jeon, Jeonghye Kim, and Youngchul Sung. Be My Tutor: On-Policy Co-Distillation for Mutual LLM Improvement via Peer Feedback. arXiv preprint arXiv:2606.14368, 2026. URL https://arxiv.org/abs/2606.14368. — [arXiv:2606.14368](https://arxiv.org/abs/2606.14368) [↗](https://arxiv.org/abs/2606.14368)
9. Yishuo Cai, Jiahui Liu, Yuanxin Liu, Haobo Deng, Linli Yao, Yuhao Zheng, Kun Ouyang, Zhimo Li, Ziyue Wang, Xu Sun, Haoli Bai, and Xiaohui Li. Thinking Without Images: Internalizing Visual Manipulation with On-Policy Self-Distillation. arXiv preprint arXiv:2606.08719, 2026a. URL [https://arxiv.org/abs…](https://arxiv.org/abs…) — [arXiv:2606.08719](https://arxiv.org/abs/2606.08719) [↗](https://arxiv.org/abs…)
10. Yuang Cai and Yuyu Yuan. X -KD: General Experiential Knowledge Distillation for Large Language Models. arXiv preprint arXiv:2602.12674, 2026. URL https://arxiv.org/abs/2602.12674. [↗](https://arxiv.org/abs/2602.12674)
11. Yuchen Cai, Ding Cao, Liang Lin, Chunxi Luo, Xin Xu, Kai Yang, Weijie Liu, Saiyong Yang, Tianxiang Zhao, Guangzhong Sun, Guiquan Liu, and Junfeng Fang. Learning to foresee: Unveiling the unlocking efficiency of on-policy distillation. arXiv preprint arXiv:2605.11739v1, 2026b. — [arXiv:2605.11739](https://arxiv.org/abs/2605.11739) [find](https://scholar.google.com/scholar?q=Yuchen+Cai%2C+Ding+Cao%2C+Liang+Lin%2C+Chunxi+Luo%2C+Xin+Xu%2C+Kai+Yang%2C+Weijie+Liu%2C+Saiyong+Yang%2C+Tianxiang+Zhao%2C+Guangzhong+Sun%2C+Guiquan+Liu%2C+and+Junfeng+Fang%2E+Learning+to+foresee%3A+Unveiling+the+unlocking+eff)
12. Di Cao, Dongjie Fu, Hai Yu, Siqi Zheng, Xu Tan, and Tao Jin. X-OPD: Cross-Modal On-Policy Distillation for Capability Alignment in Speech LLMs. arXiv preprint arXiv:2603.24596, 2026. URL https://arxiv.org/abs/2603.24596. — [arXiv:2603.24596](https://arxiv.org/abs/2603.24596) [↗](https://arxiv.org/abs/2603.24596)
13. Sungmin Cha and Kyunghyun Cho. Why Knowledge Distillation Works in Generative Models: A Minimal Working Explanation. Advances in Neural Information Processing Systems (NeurIPS), 2025. URL https://arxiv.org/abs/2505.13111. [↗](https://arxiv.org/abs/2505.13111)
14. Jingwen Chen, Wenkai Yang, Shengda Fan, Wenbo Nie, Chenxing Sun, Shaodong Zheng, Yangen Hu, Lu Pan, Ke Zeng, and Yankai Lin. Rethinking Continual Experience Internalization for Self-Evolving LLM Agents. arXiv preprint arXiv:2606.04703, 2026a. URL https://arxiv.org/abs/2606.04703. — [arXiv:2606.04703](https://arxiv.org/abs/2606.04703) [↗](https://arxiv.org/abs/2606.04703)
15. Tianlei Chen, Jiao Ou, Ziyuan Liu, Ruiming Tang, Jian Liang, and Han Li. Counteraction-Aware Multi-Teacher On-Policy Distillation for General Capability Recovery with Domain Preservation. arXiv preprint arXiv:2605.27115, 2026b. URL https://arxiv.org/abs/2605.27115. — [arXiv:2605.27115](https://arxiv.org/abs/2605.27115) [↗](https://arxiv.org/abs/2605.27115)
16. Xianwei Chen, Shimin Zhang, and Jibin Wu. f -OPD: Stabilizing Long-Horizon On-Policy Distillation with Freshness-Aware Control. arXiv preprint arXiv:2605.17862, 2026c. URL https://arxiv.org/abs/2605.17862. — [arXiv:2605.17862](https://arxiv.org/abs/2605.17862) [↗](https://arxiv.org/abs/2605.17862)
17. Xinquan Chen, Zhenyun Yin, Shan He, Bin Huang, Shanzhe Lei, Pengcheng Shi, Kun Cai, Bei Chen, Bangwei Liu, Zeyu Kang, Chao Huang, Yang Zhang, Wenjie Li, Ruijun Ge, Yajie Wang, Tianshun Fang, Tianyang Xu, Yiwen Cong, Meng Jin, Gaolei Li, Xuansheng Wu, Linhan Liu, Zijing He, An Li, Yan Teng, Xin Tan,… — [arXiv:2605.06230](https://arxiv.org/abs/2605.06230) [find](https://scholar.google.com/scholar?q=Xinquan+Chen%2C+Zhenyun+Yin%2C+Shan+He%2C+Bin+Huang%2C+Shanzhe+Lei%2C+Pengcheng+Shi%2C+Kun+Cai%2C+Bei+Chen%2C+Bangwei+Liu%2C+Zeyu+Kang%2C+Chao+Huang%2C+Yang+Zhang%2C+Wenjie+Li%2C+Ruijun+Ge%2C+Yajie+Wang%2C+Tianshun+Fang%2C+Tianyang+)
18. Zixiang Chen, Yihe Deng, Huizhuo Yuan, Kaixuan Ji, and Quanquan Gu. Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models. Proceedings of ICML, 2024. URL https://arxiv.org/abs/2401.01335. [↗](https://arxiv.org/abs/2401.01335)
19. Zizhao Chen, Yuying Li, Siting Lin, and Lianxi Wang. Efficient LLM Reasoning via Variational Posterior Guidance with Efficiency Awareness. arXiv preprint arXiv:2605.11019, 2026e. URL https://arxiv.org/abs/2605.11019. — [arXiv:2605.11019](https://arxiv.org/abs/2605.11019) [↗](https://arxiv.org/abs/2605.11019)
20. DeepSeek-AI. [DeepSeek-V4: Towards highly efficient million-token context intelligence](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2026-deepseek-v4-highly-efficient/). Technical report, DeepSeek, 2026. URL [https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main/DeepSeek](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main/DeepSeek) V4.pdf. [↗](https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main/DeepSeek)
21. DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, Bing Xue, Bingxuan Wang, Bochao Wu, Bei Feng, Chengda Lu, Chengg… — [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/)
22. Ken Ding. [HDPO: Hybrid Distillation Policy Optimization via Privileged Self-Distillation](https://md.research.elamur.ai/free/en/papers/ding-2026-hdpo-hybrid-distillation-policy-optimization/). arXiv preprint arXiv:2603.23871, 2026. URL https://arxiv.org/abs/2603.23871. [↗](https://arxiv.org/abs/2603.23871)
23. Haoyu Dong. Self-Distillation Policy Optimization via Visual Feedback: Bridging Code and Visual Artifacts. arXiv preprint arXiv:2606.10334, 2026. URL https://arxiv.org/abs/2606.10334. — [arXiv:2606.10334](https://arxiv.org/abs/2606.10334) [↗](https://arxiv.org/abs/2606.10334)
24. Jiabin Fan, Guoqing Luo, Michael Bowling, and Lili Mou. KETCHUP: K-Step Return Estimation for Sequential Knowledge Distillation. arXiv preprint arXiv:2504.19024, 2025. URL https://arxiv.org/abs/2504.19024. [↗](https://arxiv.org/abs/2504.19024)
25. Junfeng Fang, Zhepei Hong, Mao Zheng, Mingyang Song, Gengsheng Li, Houcheng Jiang, Dan Zhang, Haiyun Guo, Xiang Wang, and Tat-Seng Chua. Rubric-based On-policy Distillation. arXiv preprint arXiv:2605.07396, 2026. — [arXiv:2605.07396](https://arxiv.org/abs/2605.07396) [find](https://scholar.google.com/scholar?q=Junfeng+Fang%2C+Zhepei+Hong%2C+Mao+Zheng%2C+Mingyang+Song%2C+Gengsheng+Li%2C+Houcheng+Jiang%2C+Dan+Zhang%2C+Haiyun+Guo%2C+Xiang+Wang%2C+and+Tat%2DSeng+Chua%2E+Rubric%2Dbased+On%2Dpolicy+Distillation%2E+arXiv+preprint+arXiv%3A2605%2E)
26. Yu Fu, Longxuan Yu, Haz Sameen Shahgir, Zhipeng Wei, Hui Liu, N. Benjamin Erichson, and Yue Dong. Reducing the Safety Tax in LLM Safety Alignment with On-Policy Self-Distillation. arXiv preprint arXiv:2605.15239, 2026a. URL https://arxiv.org/abs/2605.15239. [↗](https://arxiv.org/abs/2605.15239)
27. Yuqian Fu, Haohuan Huang, Kaiwen Jiang, Jiacai Liu, Zhuo Jiang, Yuanheng Zhu, and Dongbin Zhao. [Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixes](https://md.research.elamur.ai/free/en/papers/fu-2026-revisiting-policy-distillation-empirical-failure/). arXiv preprint arXiv:2603.25562, 2026b. URL https://arxiv.org/abs/2603.25562. [↗](https://arxiv.org/abs/2603.25562)
28. Gemma Team, Morgane Riviere, Shreya Pathak, Pier Giuseppe Sessa, Cassidy Hardin, Surya Bhupatiraju, Léonard Hussenot, Thomas Mesnard, Bobak Shahriari, Alexandre Ramé, Johan Ferret, Peter Liu, Pouya Tafti, Abe Friesen, Michelle Casbon, Sabela Ramos, Ravin Kumar, Charline Le Lan, Sammy Jerome, Anton… [find](https://scholar.google.com/scholar?q=Gemma+Team%2C+Morgane+Riviere%2C+Shreya+Pathak%2C+Pier+Giuseppe+Sessa%2C+Cassidy+Hardin%2C+Surya+Bhupatiraju%2C+Le%CC%81onard+Hussenot%2C+Thomas+Mesnard%2C+Bobak+Shahriari%2C+Alexandre+Rame%CC%81%2C+Johan+Ferret%2C+Peter+Liu%2C+Pouya+)
29. Naibin Gu, Chenxu Yang, Qingyi Si, Chuanyu Qin, Dingyu Yao, Peng Fu, Zheng Lin, Weiping Wang, Nan Duan, and Jiaqi Wang. Co-Evolving Policy Distillation. arXiv preprint arXiv:2604.27083, 2026. URL https://arxiv.org/abs/2604.27083. [↗](https://arxiv.org/abs/2604.27083)
30. Yuxian Gu, Li Dong, Furu Wei, and Minlie Huang. MiniLLM: Knowledge Distillation of Large Language Models. Proceedings of ICLR, 2024. [URL https://arxiv.org/abs/2306.08543](https://md.research.elamur.ai/free/en/papers/gu-2023-minillm-policy-distillation-large-language/). [↗](https://arxiv.org/abs/2306.08543)
31. Yuxian Gu, Hao Zhou, Fandong Meng, Jie Zhou, and Minlie Huang. [MiniPLM: Knowledge Distillation for Pre-Training Language Models](https://md.research.elamur.ai/free/en/papers/gu-2024-miniplm-knowledge-distillation-pre-training/). Proceedings of ICLR, 2025. URL https://arxiv.org/abs/2410.17215. [↗](https://arxiv.org/abs/2410.17215)
32. Arnav Gudibande, Eric Wallace, Charlie Snell, Xinyang Geng, Hao Liu, Pieter Abbeel, Sergey Levine, and Dawn Song. [The False Promise of Imitating Proprietary LLMs](https://md.research.elamur.ai/free/en/papers/gudibande-2023-false-promise-imitating-proprietary-llms/). arXiv preprint arXiv:2305.15717, 2023. URL https://arxiv.org/abs/2305.15717. [↗](https://arxiv.org/abs/2305.15717)
33. Haowei Guo, Baolong Bi, Ruicheng Zhang, Bingqian Sun, and Wentao Zhang. When Should the Teacher Move? Temporal Coupling and Stability in Self On-Policy Distillation. arXiv preprint arXiv:2606.03532, 2026. URL https://arxiv.org/abs/2606.03532. [↗](https://arxiv.org/abs/2606.03532)
34. Andy Han, Kristina Fujimoto, Avidan Shah, Kiet Nguyen, Kai Xu, Chen Yueh-Han, Ilia Sucholutsky, and Rico Angell. On-Policy Consistency Training Improves LLM Safety with Minimal Capability Degradation. arXiv preprint arXiv:2605.21834, 2026a. URL https://arxiv.org/abs/2605.21834. — [arXiv:2605.21834](https://arxiv.org/abs/2605.21834) [↗](https://arxiv.org/abs/2605.21834)
35. Zihao Han, Tiangang Zhang, Huaibin Wang, and Yilun Sun. Adaptive teacher exposure for self-distillation in llm reasoning. arXiv preprint arXiv:2605.11458v1, 2026b. [find](https://scholar.google.com/scholar?q=Zihao+Han%2C+Tiangang+Zhang%2C+Huaibin+Wang%2C+and+Yilun+Sun%2E+Adaptive+teacher+exposure+for+self%2Ddistillation+in+llm+reasoning%2E+arXiv+preprint+arXiv%3A2605%2E11458v1%2C+2026b%2E)
36. Dongze Hao, Zhiwei Jin, Chen Chen, and Haonan Lu. Stabilizing On-Policy Distillation for MLLM Reasoning with Global Normalization. arXiv preprint arXiv:2606.09091, 2026. URL https://arxiv.org/abs/2606.09091. — [arXiv:2606.09091](https://arxiv.org/abs/2606.09091) [↗](https://arxiv.org/abs/2606.09091)
37. Yinghui He, Simran Kaur, Adithya Bhaskar, Yongjin Yang, Jiarui Liu, Narutatsu Ri, Liam Fowl, Abhishek Panigrahi, Danqi Chen, and Sanjeev Arora. Self-Distillation Zero: Self-Revision Turns Binary Rewards into Dense Supervision. arXiv preprint arXiv:2604.12002, 2026. URL https://arxiv.org/abs/2604.120… [↗](https://arxiv.org/abs/2604.120…)
38. Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. [Distilling the knowledge in a neural network](https://md.research.elamur.ai/free/en/papers/hinton-2015-distilling-knowledge-neural-network/). arXiv preprint arXiv:1503.02531, 2015.
39. Jordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Elena Buchatskaya, Trevor Cai, Eliza Rutherford, Diego de Las Casas, Lisa Anne Hendricks, Johannes Welbl, Aidan Clark, Tom Hennigan, Eric Noland, Katie Millican, George van den Driessche, Bogdan Damoc, Aurelia Guy, Simon Osindero, Karen Simonyan, E… — [Training Compute-Optimal Large Language Models](https://md.research.elamur.ai/free/en/papers/hoffmann-2022-training-compute-optimal-large-language/)
40. Wenjin Hou, Shangpin Peng, Weinong Wang, Zheng Ruan, Yue Zhang, Zhenglin Zhou, Mingqi Gao, Yifei Chen, Kaiqi Wang, Hongming Yang, Chengquan Zhang, Zhuotao Tian, Han Hu, Yi Yang, Fei Wu, and Hehe Fan. [Uni-opd: Unifying on-policy distillation with a dual-perspective recipe](https://md.research.elamur.ai/free/en/papers/hou-2026-uni-opd-unifying-policy-distillation/). arXiv preprint arXiv:2605.0…
41. Cheng-Yu Hsieh, Chun-Liang Li, Chih-Kuan Yeh, Hootan Nakhost, Yasuhisa Fujii, Alexander Ratner, Ranjay Krishna, Chen-Yu Lee, and Tomas Pfister. [Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes](https://md.research.elamur.ai/free/en/papers/hsieh-2023-distilling-step-step-outperforming-larger/). Findings of ACL, 2023. URL [https://arxiv.org/…](https://arxiv.org/…) [↗](https://arxiv.org/…)
42. Jian Hu, Xibin Tao, Weixun Hou, Quanzhi Liu, Zhengyu Feng, Yang Xu, Qing Ye, Qi Li, Zenan Li, and Shuai Zhang. [OpenRLHF: An easy-to-use, scalable and high-performance RLHF framework](https://md.research.elamur.ai/free/en/papers/hu-2024-openrlhf-easy-use-scalable/). arXiv preprint arXiv:2405.11143, 2024.
43. Jing Hu, Danxiang Zhu, Xianlong Luo, Dan Zhang, Shuwei He, Yishu Lei, Haitao Zheng, Shikun Feng, Jingzhou He, Yu Sun, Hua Wu, and Haifeng Wang. CORD: Bridging the Audio-Text Reasoning Gap via Weighted On-policy Cross-modal Distillation. arXiv preprint arXiv:2601.16547, 2026. URL [https://arxiv.org/ab…](https://arxiv.org/ab…) [↗](https://arxiv.org/ab…)
44. Haiduo Huang, Jiangcheng Song, Yadong Zhang, and Pengju Ren. SelecTKD: Selective Token-Weighted Knowledge Distillation for LLMs. arXiv preprint arXiv:2510.24021, 2025. URL https://arxiv.org/abs/2510.24021. — [arXiv:2510.24021](https://arxiv.org/abs/2510.24021) [↗](https://arxiv.org/abs/2510.24021)
45. Jiazhen Huang, Xiao Chen, Xiao Luo, Yong Dai, Senkang Hu, and Yuzhi Zhao. Skill-Conditioned Gated Self-Distillation for LLM Reasoning. arXiv preprint arXiv:2605.28791, 2026. URL https://arxiv.org/abs/2605.28791. [↗](https://arxiv.org/abs/2605.28791)
46. Jonas Hübotter, Frederike Lübeck, Lejs Behric, Anton Baumann, Marco Bagatella, Daniel Marta, Ido Hakimi, Idan Shenfeld, Thomas Kleine Buening, Carlos Guestrin, and Andreas Krause. [Reinforcement Learning via Self-Distillation](https://md.research.elamur.ai/free/en/papers/hubotter-2026-reinforcement-learning-self-distillation/). arXiv preprint arXiv:2601.20802, 2026. URL https://arxiv.org/abs/2601.20… [↗](https://arxiv.org/abs/2601.20…)
47. Ijun Jang, Jewon Yeom, Juan Yeo, Hyunggu Lim, and Taesup Kim. Stable On-Policy Distillation through Adaptive Target Reformulation. In Findings of ACL, 2026. URL https://arxiv.org/abs/2601.07155. [↗](https://arxiv.org/abs/2601.07155)
48. Minbyul Jeong. Healthcare AI GYM for medical agents. arXiv preprint arXiv:2605.02943, 2026. — [arXiv:2605.02943](https://arxiv.org/abs/2605.02943) [find](https://scholar.google.com/scholar?q=Minbyul+Jeong%2E+Healthcare+AI+GYM+for+medical+agents%2E+arXiv+preprint+arXiv%3A2605%2E02943%2C+2026%2E+%E2%80%94+%5BarXiv%3A2605%2E02943%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2605%2E02943%29)
49. Nan Jia, Haojin Yang, Xing Ma, Jiesong Lian, Shuailiang Zhang, Weipeng Zhang, Ke Zeng, Xunliang Cai, and Zequn Sun. [Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level](https://md.research.elamur.ai/free/en/papers/jia-2026-asymmetric-policy-distillation-bridging-exploitation/). arXiv preprint arXiv:2605.06387, 2026. URL https://arxiv.org/abs/2605.06387. [↗](https://arxiv.org/abs/2605.06387)
50. Li Jiang, Haoran Xu, Yichuan Ding, and Amy Zhang. Trajectory-Refined Distillation. arXiv preprint arXiv:2606.08432, 2026a. URL https://arxiv.org/abs/2606.08432. — [arXiv:2606.08432](https://arxiv.org/abs/2606.08432) [↗](https://arxiv.org/abs/2606.08432)
51. Yuxin Jiang, Chunkit Chan, Mingyang Chen, and Wei Wang. [Lion: Adversarial Distillation of Proprietary Large Language Models](https://md.research.elamur.ai/free/en/papers/jiang-2023-lion-adversarial-distillation-proprietary-large/). Proceedings of EMNLP, 2023. URL https://arxiv.org/abs/2305.12870. [↗](https://arxiv.org/abs/2305.12870)
52. Yuxuan Jiang and Francis Ferraro. Bridging Reasoning Trajectories in On-Policy Distillation via Near-Future Guidance. arXiv preprint arXiv:2606.00305, 2026. URL https://arxiv.org/abs/2606.00305. [↗](https://arxiv.org/abs/2606.00305)
53. Yuxuan Jiang, Runchao Li, Shubhashis Roy Dipta, Dawei Li, and Zhao Yang. Cornerstones or stumbling blocks? deciphering the rock tokens in on-policy distillation. arXiv preprint arXiv:2605.09253v1, 2026b. [find](https://scholar.google.com/scholar?q=Yuxuan+Jiang%2C+Runchao+Li%2C+Shubhashis+Roy+Dipta%2C+Dawei+Li%2C+and+Zhao+Yang%2E+Cornerstones+or+stumbling+blocks%3F+deciphering+the+rock+tokens+in+on%2Dpolicy+distillation%2E+arXiv+preprint+arXiv%3A2605%2E09253v1%2C+202)
54. Can Jin, Jiakang Li, Rui Wu, Eddy Zhang, and Dimitris N. Metaxas. Weak Critics Make Strong Learners: On-Policy Critique Distillation for Scalable Oversight. arXiv preprint arXiv:2606.00424, 2026a. URL https://arxiv.org/abs/2606.00424. — [arXiv:2606.00424](https://arxiv.org/abs/2606.00424) [↗](https://arxiv.org/abs/2606.00424)
55. Woogyeol Jin, Taywon Min, Yongjin Yang, Swanand Ravindra Kadhe, Yi Zhou, Dennis Wei, Nathalie Baracaldo, and Kimin Lee. [Entropy-Aware On-Policy Distillation of Language Models](https://md.research.elamur.ai/free/en/papers/jin-2026-entropy-aware-policy-distillation-language/). arXiv preprint arXiv:2603.07079, 2026b. URL https://arxiv.org/abs/2603.07079. [↗](https://arxiv.org/abs/2603.07079)
56. Yiqiao Jin, Yiyang Wang, Lucheng Fu, Yijia Xiao, Yinyi Luo, Haoxin Liu, B. Aditya Prakash, Josiah Hester, Jindong Wang, and Srijan Kumar. UniSD: Towards a Unified Self-Distillation Framework for Large Language Models. arXiv preprint arXiv:2605.06597, 2026c. [find](https://scholar.google.com/scholar?q=Yiqiao+Jin%2C+Yiyang+Wang%2C+Lucheng+Fu%2C+Yijia+Xiao%2C+Yinyi+Luo%2C+Haoxin+Liu%2C+B%2E+Aditya+Prakash%2C+Josiah+Hester%2C+Jindong+Wang%2C+and+Srijan+Kumar%2E+UniSD%3A+Towards+a+Unified+Self%2DDistillation+Framework+for+Large)
57. Jaehun Jung, Hyunwoo Kim, Brandon Cui, Ximing Lu, David Acuna, Prithviraj Ammanabrolu, and Yejin Choi. DeltaPrompts: Escaping the Zero-Delta Trap in Multimodal Distillation. arXiv preprint arXiv:2605.15532, 2026. URL https://arxiv.org/abs/2605.15532. [↗](https://arxiv.org/abs/2605.15532)
58. Jared Kaplan, Sam McCandlish, Tom Henighan, Tom B Brown, Benjamin Chess, Rewon Child, Scott Gray, Alec Radford, Jeffrey Wu, and Dario Amodei. [Scaling laws for neural language models](https://md.research.elamur.ai/free/en/papers/kaplan-2020-scaling-laws-neural-language-models/). arXiv preprint arXiv:2001.08361, 2020.
59. Junlong Ke, Yuxin Huang, Jingdi Lei, Zefan Cai, Xiangyu Peng, and Baobao Chang. Respecting self-uncertainty in on-policy self-distillation for efficient LLM reasoning. arXiv preprint arXiv:2605.13255, 2026. — [arXiv:2605.13255](https://arxiv.org/abs/2605.13255) [find](https://scholar.google.com/scholar?q=Junlong+Ke%2C+Yuxin+Huang%2C+Jingdi+Lei%2C+Zefan+Cai%2C+Xiangyu+Peng%2C+and+Baobao+Chang%2E+Respecting+self%2Duncertainty+in+on%2Dpolicy+self%2Ddistillation+for+efficient+LLM+reasoning%2E+arXiv+preprint+arXiv%3A2605%2E13255%2C)
60. Fatemeh Khadem, Sajad Mousavi, Yi Fang, and Yuhong Liu. DP-OPD: Differentially Private On-Policy Distillation for Language Models. arXiv preprint arXiv:2604.04461, 2026. URL https://arxiv.org/abs/2604.04461. — [arXiv:2604.04461](https://arxiv.org/abs/2604.04461) [↗](https://arxiv.org/abs/2604.04461)
61. Gyeongman Kim, Doohyuk Jang, and Eunho Yang. PromptKD: Distilling Student-Friendly Knowledge for Generative Language Models via Prompt Tuning. In Findings of EMNLP, 2024. URL https://arxiv.org/abs/2402.12842. [↗](https://arxiv.org/abs/2402.12842)
62. Jaehoon Kim and Dongha Lee. OPSD Compresses What RLVR Teaches: A Post-RL Compaction Stage for Reasoning Models. arXiv preprint arXiv:2605.06188, 2026. URL https://arxiv.org/abs/2605.06188. — [arXiv:2605.06188](https://arxiv.org/abs/2605.06188) [↗](https://arxiv.org/abs/2605.06188)
63. Jeonghye Kim, Jiwon Jeon, Dongsheng Li, and Yuqing Yang. Rebellious Student: Reversing Teacher Signals for Reasoning Exploration with Self-Distilled RLVR. arXiv preprint arXiv:2605.10781, 2026a. — [arXiv:2605.10781](https://arxiv.org/abs/2605.10781) [find](https://scholar.google.com/scholar?q=Jeonghye+Kim%2C+Jiwon+Jeon%2C+Dongsheng+Li%2C+and+Yuqing+Yang%2E+Rebellious+Student%3A+Reversing+Teacher+Signals+for+Reasoning+Exploration+with+Self%2DDistilled+RLVR%2E+arXiv+preprint+arXiv%3A2605%2E10781%2C+2026a%2E+%E2%80%94+%5Bar)
64. Jeonghye Kim, Xufang Luo, Minbeom Kim, Sangmook Lee, Dohyung Kim, Jiwon Jeon, Dongsheng Li, and Yuqing Yang. Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs? arXiv preprint arXiv:2603.24472, 2026b. URL https://arxiv.org/abs/2603.24472. — [arXiv:2603.24472](https://arxiv.org/abs/2603.24472) [↗](https://arxiv.org/abs/2603.24472)
65. Minsang Kim and Seung Jun Baek. [Explain in Your Own Words: Improving Reasoning via Token-Selective Dual Knowledge Distillation](https://md.research.elamur.ai/free/en/papers/kim-2026-explain-own-words-improving-reasoning/). Proceedings of ICLR, 2026. URL https://arxiv.org/abs/2603.13260. [↗](https://arxiv.org/abs/2603.13260)
66. Yeongmin Kim, Donghyeok Shin, Mina Kang, Byeonghu Na, and Il-Chul Moon. Distillation of Large Language Models via Concrete Score Matching. In International Conference on Learning Representations (ICLR), 2026c. URL https://arxiv.org/abs/2509.25837. [↗](https://arxiv.org/abs/2509.25837)
67. Yoon Kim and Alexander M. Rush. [Sequence-level knowledge distillation](https://md.research.elamur.ai/free/en/papers/kim-2016-sequence-level-knowledge-distillation/). In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pp. 1317–1327, 2016.
68. John Kirchenbauer, Abhimanyu Hans, Brian Bartoldson, Micah Goldblum, Ashwinee Panda, and Tom Goldstein. Multi-Token Prediction via Self-Distillation. arXiv preprint arXiv:2602.06019, 2026. URL https://arxiv.org/abs/2602.06019. — [arXiv:2602.06019](https://arxiv.org/abs/2602.06019) [↗](https://arxiv.org/abs/2602.06019)
69. James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, et al. Overcoming catastrophic forgetting in neural networks. Proceedings of the National Academy of Sciences, 114(13):3521–3526,… [find](https://scholar.google.com/scholar?q=James+Kirkpatrick%2C+Razvan+Pascanu%2C+Neil+Rabinowitz%2C+Joel+Veness%2C+Guillaume+Desjardins%2C+Andrei+A+Rusu%2C+Kieran+Milan%2C+John+Quan%2C+Tiago+Ramalho%2C+Agnieszka+Grabska%2DBarwinska%2C+et+al%2E+Overcoming+catastrophi)
70. Jongwoo Ko, Sungnyun Kim, Tianyi Chen, and Se-Young Yun. [DistiLLM: Towards Streamlined Distillation for Large Language Models](https://md.research.elamur.ai/free/en/papers/ko-2024-distillm-towards-streamlined-distillation-large/). Proceedings of ICML, 2024. URL https://arxiv.org/abs/2402.03898. [↗](https://arxiv.org/abs/2402.03898)
71. Jongwoo Ko, Tianyi Chen, Sungnyun Kim, Tianyu Ding, Luming Liang, Ilya Zharkov, and Se-Young Yun. [DistiLLM-2: A Contrastive Approach Boosts the Distillation of LLMs](https://md.research.elamur.ai/free/en/papers/ko-2025-distillm-2-contrastive-approach-boosts/). In Proceedings of ICML, 2025. URL https://arxiv.org/abs/2503.07067. [↗](https://arxiv.org/abs/2503.07067)
72. Jongwoo Ko, Sara Abdali, Young Jin Kim, Tianyi Chen, and Pashmina Cameron. [Scaling Reasoning Efficiently via Relaxed On-Policy Distillation](https://md.research.elamur.ai/free/en/papers/ko-2026-scaling-reasoning-efficiently-relaxed-policy/). arXiv preprint arXiv:2603.11137, 2026. URL https://arxiv.org/abs/2603.11137. [↗](https://arxiv.org/abs/2603.11137)
73. Avinash Kumar, Sujay Sanghavi, and Poulami Das. Test-time speculation. arXiv preprint arXiv:2605.09329v1, 2026. URL https://arxiv.org/abs/2605.09329. — [arXiv:2605.09329](https://arxiv.org/abs/2605.09329) [↗](https://arxiv.org/abs/2605.09329)
74. Soo Min Kwon, Ziteng Sun, Ananda Theertha Suresh, Himanshu Jain, and Sanjiv Kumar. Codistill-grpo: A co-distillation recipe for efficient group relative policy optimization. arXiv preprint arXiv:2605.08873v1, 2026. — [arXiv:2605.08873](https://arxiv.org/abs/2605.08873) [find](https://scholar.google.com/scholar?q=Soo+Min+Kwon%2C+Ziteng+Sun%2C+Ananda+Theertha+Suresh%2C+Himanshu+Jain%2C+and+Sanjiv+Kumar%2E+Codistill%2Dgrpo%3A+A+co%2Ddistillation+recipe+for+efficient+group+relative+policy+optimization%2E+arXiv+preprint+arXiv%3A2605%2E)
75. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph Gonzalez, Hao Zhang, and Ion Stoica. [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 ACM SIGOPS 29th Symposium on Operating Systems Principles (SOSP), 2023.
76. Changhao Li, Bohao Yang, Zexian Li, Chenyang Zhao, Jiahao Li, Hongjin Su, Qian Liu, and Tao Yu. Revisiting DAgger in the era of LLM-agents. arXiv preprint arXiv:2605.12913, 2026a. — [arXiv:2605.12913](https://arxiv.org/abs/2605.12913) [find](https://scholar.google.com/scholar?q=Changhao+Li%2C+Bohao+Yang%2C+Zexian+Li%2C+Chenyang+Zhao%2C+Jiahao+Li%2C+Hongjin+Su%2C+Qian+Liu%2C+and+Tao+Yu%2E+Revisiting+DAgger+in+the+era+of+LLM%2Dagents%2E+arXiv+preprint+arXiv%3A2605%2E12913%2C+2026a%2E+%E2%80%94+%5BarXiv%3A2605%2E12913%5D)
77. Fengxiang Li, Han Zhang, Haoyang Huang, Jinghui Wang, Jinhua Hao, Kun Yuan, Mengtong Li, Minglei Zhang, Pengcheng Xu, Wenhao Zhuang, Yizhen Shao, Zongxian Feng, Can Tang, Chao Wang, Chengxiao Tong, Fan Yang, Gang Xiong, Haixuan Gao, Han Gao, Hao Wang, Haochen Liu, Hongliang Sun, Jiabao Li, Jingwen C… — [arXiv:2603.27703](https://arxiv.org/abs/2603.27703) [find](https://scholar.google.com/scholar?q=Fengxiang+Li%2C+Han+Zhang%2C+Haoyang+Huang%2C+Jinghui+Wang%2C+Jinhua+Hao%2C+Kun+Yuan%2C+Mengtong+Li%2C+Minglei+Zhang%2C+Pengcheng+Xu%2C+Wenhao+Zhuang%2C+Yizhen+Shao%2C+Zongxian+Feng%2C+Can+Tang%2C+Chao+Wang%2C+Chengxiao+Tong%2C+Fa)
78. Gengsheng Li, Tianyu Yang, Junfeng Fang, Mingyang Song, Mao Zheng, Haiyun Guo, Dan Zhang, Jinqiao Wang, and Tat-Seng Chua. Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing. arXiv preprint arXiv:2604.02288, 2026c. URL https://arxiv.org/abs/2604.02288. — [arXiv:2604.02288](https://arxiv.org/abs/2604.02288) [↗](https://arxiv.org/abs/2604.02288)
79. Guankai Li, Jiabin Chen, Yi Xu, Xichen Zhang, and Yuan Lu. Hypereyes: Dual-grained efficiency-aware reinforcement learning for parallel multimodal search agents. arXiv preprint arXiv:2605.07177v2, 2026d. URL https://arxiv.org/abs/2605.07177. — [arXiv:2605.07177](https://arxiv.org/abs/2605.07177) [↗](https://arxiv.org/abs/2605.07177)
80. Hao Li, Jingkun An, Zijun Song, Pengyu Zhu, Rui Li, Hao Wang, Wendi Feng, Yesheng Liu, Lijun Li, Jin-Ge Yao, and Lei Sha. SafeSteer: Localized On-Policy Distillation for Efficient Safety Alignment. arXiv preprint arXiv:2606.02530, 2026e. URL https://arxiv.org/abs/2606.02530. — [arXiv:2606.02530](https://arxiv.org/abs/2606.02530) [↗](https://arxiv.org/abs/2606.02530)
81. Jiaze Li, Hao Yin, Haoran Xu, Boshen Xu, Wenhui Tan, Zewen He, Jianzhong Ju, Zhenbo Luo, and Jian Luan. Video-OPD: Efficient Post-Training of Multimodal Large Language Models for Temporal Video Grounding via On-Policy Distillation. arXiv preprint arXiv:2602.02994, 2026f. URL https://arxiv.org/abs/26… — [arXiv:2602.02994](https://arxiv.org/abs/2602.02994) [↗](https://arxiv.org/abs/26…)
82. Sijia Li, Jiazuo Yu, Zhoujun Li, Buyun Zhang, Reza Azad, Salman Khan, and Fahad Shahbaz Khan. GEAR: Granularity-adaptive advantage reweighting for LLM agents via self-distillation. arXiv preprint arXiv:2605.11853, 2026g. — [arXiv:2605.11853](https://arxiv.org/abs/2605.11853) [find](https://scholar.google.com/scholar?q=Sijia+Li%2C+Jiazuo+Yu%2C+Zhoujun+Li%2C+Buyun+Zhang%2C+Reza+Azad%2C+Salman+Khan%2C+and+Fahad+Shahbaz+Khan%2E+GEAR%3A+Granularity%2Dadaptive+advantage+reweighting+for+LLM+agents+via+self%2Ddistillation%2E+arXiv+preprint+arXi)
83. Xin Li, Hao Jiang, Annan Wang, Yichi Zhang, and Chau Yuen. The extrapolation cliff in on-policy distillation of near-deterministic structured outputs. arXiv preprint arXiv:2605.08737v1, 2026h. — [arXiv:2605.08737](https://arxiv.org/abs/2605.08737) [find](https://scholar.google.com/scholar?q=Xin+Li%2C+Hao+Jiang%2C+Annan+Wang%2C+Yichi+Zhang%2C+and+Chau+Yuen%2E+The+extrapolation+cliff+in+on%2Dpolicy+distillation+of+near%2Ddeterministic+structured+outputs%2E+arXiv+preprint+arXiv%3A2605%2E08737v1%2C+2026h%2E+%E2%80%94+%5BarXi)
84. Yang Li, Erik Nijkamp, Semih Yavuz, and Shafiq Rayhan Joty. Learning from language feedback via variational policy distillation. arXiv preprint arXiv:2605.15113, 2026i. URL https://arxiv.org/abs/2605.15113. [↗](https://arxiv.org/abs/2605.15113)
85. Yaxuan Li, Yuxin Zuo, Bingxiang He, Jinqian Zhang, Chaojun Xiao, Cheng Qian, Tianyu Yu, Huan ang Gao, Wenkai Yang, Zhiyuan Liu, and Ning Ding. [Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe](https://md.research.elamur.ai/free/en/papers/li-2026-rethinking-policy-distillation-large-language/). arXiv preprint arXiv:2604.13016, 2026j. URL [https://arxiv.o…](https://arxiv.o…) [↗](https://arxiv.o…)
86. Yingru Li, Ziniu Li, and Jiacai Liu. A Note on Hybrid Online Reinforcement and Imitation Learning for LLMs: Formulations and Algorithms. arXiv preprint arXiv:2512.23097, 2025. URL https://arxiv.org/abs/2512.23097. [↗](https://arxiv.org/abs/2512.23097)
87. Yu Li, Rui Miao, Tian Lan, and Zhengling Qi. OPPO: Bayesian Value Recursion for Token-Level Credit Assignment in LLM Reasoning. arXiv preprint arXiv:2605.21851, 2026k. URL https://arxiv.org/abs/2605.21851. — [arXiv:2605.21851](https://arxiv.org/abs/2605.21851) [↗](https://arxiv.org/abs/2605.21851)
88. Yuduo Li, Xiaofeng Shi, Qian Kou, Longbin Yu, and Hua Zhou. RAFT: Data Refinement and Adaptive Distillation for Domain Fine-Tuning with Alleviated Forgetting. arXiv preprint arXiv:2606.00147, 2026l. URL https://arxiv.org/abs/2606.00147. — [arXiv:2606.00147](https://arxiv.org/abs/2606.00147) [↗](https://arxiv.org/abs/2606.00147)
89. Yuying Li, Leqi Zheng, Yongzi Yu, Wenrui Zhou, Xuchang Zhong, Xing Hu, Jing Jin, Hangjie Yuan, and Tao Feng. Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation. arXiv preprint arXiv:2606.02684, 2026m. URL https://arxiv.org/abs/2606.02684. — [arXiv:2606.02684](https://arxiv.org/abs/2606.02684) [↗](https://arxiv.org/abs/2606.02684)
90. Kun Liang, Clive Bai, Xin Xu, Chenming Tang, Sanwoo Lee, Weijie Liu, Saiyong Yang, and Yunfang Wu. ORBIT: On-policy Exploration-Exploitation for Controllable Multi-Budget Reasoning. arXiv preprint arXiv:2601.08310, 2026a. URL https://arxiv.org/abs/2601.08310. — [arXiv:2601.08310](https://arxiv.org/abs/2601.08310) [↗](https://arxiv.org/abs/2601.08310)
91. Percy Liang, Rishi Bommasani, Tony Lee, Dimitris Tsipras, Dilara Soylu, Michihiro Yasunaga, Yian Zhang, Deepak Narayanan, Yuhuai Wu, Ananya Kumar, Benjamin Newman, Binhang Yuan, Bobby Yan, Ce Zhang, Christian Cosgrove, Christopher D. Manning, Christopher Ré, Diana Acosta-Navas, Drew A. Hudson, Eric… [find](https://scholar.google.com/scholar?q=Percy+Liang%2C+Rishi+Bommasani%2C+Tony+Lee%2C+Dimitris+Tsipras%2C+Dilara+Soylu%2C+Michihiro+Yasunaga%2C+Yian+Zhang%2C+Deepak+Narayanan%2C+Yuhuai+Wu%2C+Ananya+Kumar%2C+Benjamin+Newman%2C+Binhang+Yuan%2C+Bobby+Yan%2C+Ce+Zhang%2C+C)
92. Zihan Liang, Yufei Ma, Ben Chen, Zhipeng Qian, Xuxin Zhang, Huangyu Dai, and Lingtao Mao. Search-E1: Self-Distillation Drives Self-Evolution in Search-Augmented Reasoning. arXiv preprint arXiv:2605.22511, 2026b. URL https://arxiv.org/abs/2605.22511. [↗](https://arxiv.org/abs/2605.22511)
93. Wenjie Liao, Like Wu, Liangjie Zhao, Shihui Xu, and Shigeru Fujimura. IRIS: Interpolative Rényi Iterative Self-play for Large Language Model Fine-Tuning. arXiv preprint arXiv:2604.20933, 2026. URL https://arxiv.org/abs/2604.20933. — [arXiv:2604.20933](https://arxiv.org/abs/2604.20933) [↗](https://arxiv.org/abs/2604.20933)
94. Hunter Lightman, Vineet Kosaraju, Yura Burda, Harri Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. Let’s Verify Step by Step. Proceedings of ICLR, 2024. [URL https://arxiv.org/abs/2305.20050](https://md.research.elamur.ai/free/en/papers/lightman-2023-let-s-verify-step-step/). [↗](https://arxiv.org/abs/2305.20050)
95. Hao Lin, Kunyang Lv, Xu Jiang, Jingqi Tian, Zhongjing Du, Jiayu Ding, Qiaoman Zhang, and Hongbo Jin. VISD: Enhancing Video Reasoning via Structured Self-Distillation. arXiv preprint arXiv:2605.06094, 2026a. URL https://arxiv.org/abs/2605.06094. — [arXiv:2605.06094](https://arxiv.org/abs/2605.06094) [↗](https://arxiv.org/abs/2605.06094)
96. Wenye Lin and Kai Han. Surgical Post-Training: Proximal On-Policy Distillation for Reasoning with Knowledge Retention. arXiv preprint arXiv:2603.01683, 2026. URL https://arxiv.org/abs/2603.01683. [↗](https://arxiv.org/abs/2603.01683)
97. Zizhuo Lin, Quanling Liu, Jinsheng Quan, Chao Zhang, Yifan Zhu, Xing Shi, Jingtao Xu, Zhihui Li, and Yawei Luo. Same Evidence, Different Answers: Canonical-Context On-Policy Distillation for Multi-Turn Language Models. arXiv preprint arXiv:2605.30251, 2026b. URL https://arxiv.org/abs/2605.30251. — [arXiv:2605.30251](https://arxiv.org/abs/2605.30251) [↗](https://arxiv.org/abs/2605.30251)
98. Haoran Liu, Yuwei Zhang, Xiyao Li, Bohan Lyu, and Jingbo Shang. HERO: Hindsight-Enhanced Reflection from Environment Observations for Agentic Self-Distillation. arXiv preprint arXiv:2606.11559, 2026a. URL https://arxiv.org/abs/2606.11559. — [arXiv:2606.11559](https://arxiv.org/abs/2606.11559) [↗](https://arxiv.org/abs/2606.11559)
99. Kaiyuan Liu, Ruijie Xu, Yifan Zhang, Silu Mao, Lizhen Qu, Shuming Shi, and Zhi-Hong Deng. Prefix teach, suffix fade: Local teachability collapse in strong-to-weak on-policy distillation. arXiv preprint arXiv:2605.13643, 2026b. — [arXiv:2605.13643](https://arxiv.org/abs/2605.13643) [find](https://scholar.google.com/scholar?q=Kaiyuan+Liu%2C+Ruijie+Xu%2C+Yifan+Zhang%2C+Silu+Mao%2C+Lizhen+Qu%2C+Shuming+Shi%2C+and+Zhi%2DHong+Deng%2E+Prefix+teach%2C+suffix+fade%3A+Local+teachability+collapse+in+strong%2Dto%2Dweak+on%2Dpolicy+distillation%2E+arXiv+preprin)
100. Ruiqi Liu, Xiaolei Lv, Gengsheng Li, Ximo Zhu, Zhiheng Wang, Zhengbo Zhang, Junkai Chen, Zhiheng Li, Bo Li, Jun Gao, and Shu Wu. Visual-Advantage On-Policy Distillation for Vision-Language Models. arXiv preprint arXiv:2605.21924, 2026c. URL https://arxiv.org/abs/2605.21924. — [arXiv:2605.21924](https://arxiv.org/abs/2605.21924) [↗](https://arxiv.org/abs/2605.21924)
101. Xiaogeng Liu, Xinyan Wang, Yingzi Ma, Yechao Zhang, and Chaowei Xiao. [When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning](https://md.research.elamur.ai/free/en/papers/liu-2026-teacher-tokens-reliable-position-weighted/). arXiv preprint arXiv:2605.21606, 2026d. URL https://arxiv.org/abs/2605.21606. [↗](https://arxiv.org/abs/2605.21606)
102. Xinyu Liu, Kechen Jiao, Chunyang Xiao, Runsong Zhao, Junhao Ruan, Bei Li, Jiahao Liu, Qifan Wang, Xin Chen, Jingang Wang, Tong Xiao, and JingBo Zhu. Teacher-guided policy optimization for LLM distillation. arXiv preprint arXiv:2605.13230, 2026e. — [arXiv:2605.13230](https://arxiv.org/abs/2605.13230) [find](https://scholar.google.com/scholar?q=Xinyu+Liu%2C+Kechen+Jiao%2C+Chunyang+Xiao%2C+Runsong+Zhao%2C+Junhao+Ruan%2C+Bei+Li%2C+Jiahao+Liu%2C+Qifan+Wang%2C+Xin+Chen%2C+Jingang+Wang%2C+Tong+Xiao%2C+and+JingBo+Zhu%2E+Teacher%2Dguided+policy+optimization+for+LLM+distilla)
103. Yanjiang Liu, Jie Lou, Xinyan Guan, Yuqiu Ji, Hongyu Lin, Ben He, Xianpei Han, Le Sun, Xing Yu, and Yaojie Lu. Your Teacher Can’t Help You Here: Combating Supervision Fidelity Decay in On-Policy Distillation. arXiv preprint arXiv:2605.30833, 2026f. URL https://arxiv.org/abs/2605.30833. — [arXiv:2605.30833](https://arxiv.org/abs/2605.30833) [↗](https://arxiv.org/abs/2605.30833)
104. Yifeng Liu, Shiyuan Zhang, Yifan Zhang, and Quanquan Gu. [Self-Distilled Policy Gradient](https://md.research.elamur.ai/free/en/papers/liu-2026-self-distilled-policy-gradient/). arXiv preprint arXiv:2606.04036, 2026g. URL https://arxiv.org/abs/2606.04036. [↗](https://arxiv.org/abs/2606.04036)
105. Yihong Liu, Raoyuan Zhao, Michael A. Hedderich, and Hinrich Schütze. Crosslingual on-policy self-distillation for multilingual reasoning. arXiv preprint arXiv:2605.09548v1, 2026h. — [arXiv:2605.09548](https://arxiv.org/abs/2605.09548) [find](https://scholar.google.com/scholar?q=Yihong+Liu%2C+Raoyuan+Zhao%2C+Michael+A%2E+Hedderich%2C+and+Hinrich+Schu%CC%88tze%2E+Crosslingual+on%2Dpolicy+self%2Ddistillation+for+multilingual+reasoning%2E+arXiv+preprint+arXiv%3A2605%2E09548v1%2C+2026h%2E+%E2%80%94+%5BarXiv%3A2605%2E09548)
106. Youwei Liu, Jian Wang, Hanlin Wang, and Wenjie Li. COMAP: Co-Evolving World Models and Agent Policies for LLM Agents. arXiv preprint arXiv:2606.02372, 2026i. URL https://arxiv.org/abs/2606.02372. — [arXiv:2606.02372](https://arxiv.org/abs/2606.02372) [↗](https://arxiv.org/abs/2606.02372)
107. Kevin Lu and Thinking Machines Lab. On-policy distillation. Thinking Machines Lab: Connectionism, 2025. URL [https://www.thinkingmachines.ai/blog/on-policy-distillation](https://www.thinkingmachines.ai/blog/on-policy-distillation). doi: 10.64434/tml.20251026. — [doi:10.64434/tml.20251026](https://doi.org/10.64434/tml.20251026) [↗](https://www.thinkingmachines.ai/blog/on-policy-distillation)
108. Zhengxi Lu, Zhiyuan Yao, Zhuowen Han, Zi-Han Wang, Jinyang Wu, Qi Gu, Xunliang Cai, Weiming Lu, Jun Xiao, Yueting Zhuang, and Yongliang Shen. Self-distilled agentic reinforcement learning. arXiv preprint arXiv:2605.15155, 2026. — [arXiv:2605.15155](https://arxiv.org/abs/2605.15155) [find](https://scholar.google.com/scholar?q=Zhengxi+Lu%2C+Zhiyuan+Yao%2C+Zhuowen+Han%2C+Zi%2DHan+Wang%2C+Jinyang+Wu%2C+Qi+Gu%2C+Xunliang+Cai%2C+Weiming+Lu%2C+Jun+Xiao%2C+Yueting+Zhuang%2C+and+Yongliang+Shen%2E+Self%2Ddistilled+agentic+reinforcement+learning%2E+arXiv+prepr)
109. Feng Luo, Yu-Neng Chuang, Guanchu Wang, Zicheng Xu, Xiaotian Han, Tianyi Zhang, and Vladimir Braverman. Demystifying OPD: Length Inflation and Stabilization Strategies for Large Language Models. arXiv preprint arXiv:2604.08527, 2026a. URL https://arxiv.org/abs/2604.08527. — [arXiv:2604.08527](https://arxiv.org/abs/2604.08527) [↗](https://arxiv.org/abs/2604.08527)
110. Haocheng Luo, Jiahui Liu, Ruicheng Zhang, Zhizhou Zhong, Jiaqi Huang, Zunnan Xu, Quan Shi, Jun Zhou, and Xiu Li. Learning Visual Spatial Planning from Symbolic State via Modality-Gap-Aware Self-Distillation. arXiv preprint arXiv:2606.06076, 2026b. URL https://arxiv.org/abs/2606.06076. — [arXiv:2606.06076](https://arxiv.org/abs/2606.06076) [↗](https://arxiv.org/abs/2606.06076)
111. Yuanjie Lyu, Chengyu Wang, Jun Huang, and Tong Xu. From Correction to Mastery: Reinforced Distillation of Large Language Model Agents. arXiv preprint arXiv:2509.14257, 2025. URL https://arxiv.org/abs/2509.14257. — [arXiv:2509.14257](https://arxiv.org/abs/2509.14257) [↗](https://arxiv.org/abs/2509.14257)
112. Yufei Ma, Zihan Liang, Ben Chen, Zhipeng Qian, Huangyu Dai, Lingtao Mao, Xuxin Zhang, Chenyi Lei, and Wenwu Ou. SD-Search: On-Policy Hindsight Self-Distillation for Search-Augmented Reasoning. arXiv preprint arXiv:2605.18299, 2026. URL https://arxiv.org/abs/2605.18299. [↗](https://arxiv.org/abs/2605.18299)
113. Duy Nguyen, Hanqi Xiao, Archiki Prasad, Zaid Khan, Anirban Das, Austin Zhang, Sambit Sahu, Hyunji Lee, Elias Stengel-Eskin, and Mohit Bansal. AVSD: Adaptive-View Self-Distillation by Balancing Consensus and Teacher-Specific Privileged Signals. arXiv preprint arXiv:2605.20643, 2026. URL https://arxiv… — [arXiv:2605.20643](https://arxiv.org/abs/2605.20643) [find](https://scholar.google.com/scholar?q=Duy+Nguyen%2C+Hanqi+Xiao%2C+Archiki+Prasad%2C+Zaid+Khan%2C+Anirban+Das%2C+Austin+Zhang%2C+Sambit+Sahu%2C+Hyunji+Lee%2C+Elias+Stengel%2DEskin%2C+and+Mohit+Bansal%2E+AVSD%3A+Adaptive%2DView+Self%2DDistillation+by+Balancing+Consens)
114. Yifan Niu, Han Xiao, Dongyi Liu, Zelong Wang, Dihong Gong, Yasheng Wang, and Jia Li. Breaking the Tokenizer Barrier: On-Policy Distillation across Model Families. arXiv preprint arXiv:2606.09456, 2026. URL https://arxiv.org/abs/2606.09456. [↗](https://arxiv.org/abs/2606.09456)
115. NVIDIA. TensorRT-LLM: High-performance inference framework for large language models. [https://github.com/NVIDIA/TensorRT-LLM](https://github.com/NVIDIA/TensorRT-LLM), 2024. [↗](https://github.com/NVIDIA/TensorRT-LLM)
116. Minjae Oh, Sangjun Song, Gyubin Choi, Yunho Choi, and Yohan Jo. [KL for a KL: On-Policy Distillation with Control Variate Baseline](https://md.research.elamur.ai/free/en/papers/oh-2026-kl-kl-policy-distillation-control/). arXiv preprint arXiv:2605.07865, 2026.
117. Leyi Pan, Shuchang Tao, Yunpeng Zhai, Lingzhe Zhang, Zhaoyang Liu, Bolin Ding, Aiwei Liu, and Lijie Wen. RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation. arXiv preprint arXiv:2606.11709, 2026. URL https://arxiv.org/abs/2606.11709. — [arXiv:2606.11709](https://arxiv.org/abs/2606.11709) [↗](https://arxiv.org/abs/2606.11709)
118. Sangwoo Park, Woongyeong Yeo, Seanie Lee, Yumin Choi, Hyomin Lee, Kangsan Kim, Jinheon Baek, Seong Joon Oh, and Sung Ju Hwang. It Takes Two: Complementary Self-Distillation for Contextual Integrity in LLMs. arXiv preprint arXiv:2605.20258, 2026. URL https://arxiv.org/abs/2605.20258. [↗](https://arxiv.org/abs/2605.20258)
119. Patomporn Payoungkhamdee, Tinnakit Udsa, Jian Gang Ngui, Sarana Nutanong, Alham Fikri Aji, and Peerat Limkonchotiwat. DuDi: Dual-Signal Distillation with Cross-Lingual Verbalizer. arXiv preprint arXiv:2606.04694, 2026. URL https://arxiv.org/abs/2606.04694. [↗](https://arxiv.org/abs/2606.04694)
120. Emiliano Penaloza, Dheeraj Vattikonda, Nicolas Gontier, Alexandre Lacoste, Laurent Charlin, and Massimo Caccia. Privileged Information Distillation for Language Models. arXiv preprint arXiv:2602.04942, 2026. URL https://arxiv.org/abs/2602.04942. — [arXiv:2602.04942](https://arxiv.org/abs/2602.04942) [↗](https://arxiv.org/abs/2602.04942)
121. Jingyu Peng, Maolin Wang, Hengyi Cai, Yuchen Li, Kai Zhang, Shuaiqiang Wang, Dawei Yin, and Xiangyu Zhao. AdaSwitch: Balancing Exploration and Guidance in Knowledge Distillation via Adaptive Switching. arXiv preprint arXiv:2510.07842, 2025. URL https://arxiv.org/abs/2510.07842. — [arXiv:2510.07842](https://arxiv.org/abs/2510.07842) [↗](https://arxiv.org/abs/2510.07842)
122. Daniil Plyusov, Alexey Gorbatovski, Alexey Malakhov, Nikita Balagansky, Boris Shaposhnikov, Daria Korotyshova, and Daniil Gavrilov. Trust-Region Behavior Blending for On-Policy Distillation. arXiv preprint arXiv:2605.31159, 2026. URL https://arxiv.org/abs/2605.31159. [↗](https://arxiv.org/abs/2605.31159)
123. Ruiyang Qin, Qingzhuo Wang, Dongrui Liu, Qiang Li, Zhihua Wei, and Wen Shen. Multilingual Safety Alignment via Self-Distillation. arXiv preprint arXiv:2605.02971, 2026. URL https://arxiv.org/abs/2605.02971. — [arXiv:2605.02971](https://arxiv.org/abs/2605.02971) [↗](https://arxiv.org/abs/2605.02971)
124. Miguel Moura Ramos, Duarte M. Alves, and André F. T. Martins. Combining on-policy optimization and distillation for long-context reasoning in large language models. arXiv preprint arXiv:2605.12227v1, 2026. — [arXiv:2605.12227](https://arxiv.org/abs/2605.12227) [find](https://scholar.google.com/scholar?q=Miguel+Moura+Ramos%2C+Duarte+M%2E+Alves%2C+and+Andre%CC%81+F%2E+T%2E+Martins%2E+Combining+on%2Dpolicy+optimization+and+distillation+for+long%2Dcontext+reasoning+in+large+language+models%2E+arXiv+preprint+arXiv%3A2605%2E12227v1%2C)
125. Miao Rang, Zhenni Bi, Hang Zhou, Kai Han, Xuechun Wang, An Xiao, Xinghao Chen, Yunhe Wang, and Hanting Chen. Near-Policy: Accelerating On-Policy Distillation via Asynchronous Generation and Selective Packing. arXiv preprint arXiv:2605.05940, 2026. URL https://arxiv.org/abs/2605.05940. — [arXiv:2605.05940](https://arxiv.org/abs/2605.05940) [↗](https://arxiv.org/abs/2605.05940)
126. MohammadHossein Rezaei, Anas Mahmoud, Zihao Wang, Utkarsh Tyagi, Advait Gosai, Razvan-Gabriel Dumitru, Aakash Sabharwal, Bing Liu, and Yunzhong He. Rubric-Guided Self-Distillation: Post-Training Without Rubric Verifiers. arXiv preprint arXiv:2606.12507, 2026. URL https://arxiv.org/abs/2606.12507. [↗](https://arxiv.org/abs/2606.12507)
127. Stéphane Ross, Geoffrey J. Gordon, and J. Andrew Bagnell. A reduction of imitation learning and structured prediction to no-regret online learning. In Proceedings of the 14th International Conference on Artificial Intelligence and Statistics, pp. 627–635, 2011. [find](https://scholar.google.com/scholar?q=Ste%CC%81phane+Ross%2C+Geoffrey+J%2E+Gordon%2C+and+J%2E+Andrew+Bagnell%2E+A+reduction+of+imitation+learning+and+structured+prediction+to+no%2Dregret+online+learning%2E+In+Proceedings+of+the+14th+International+Conference)
128. Hejian Sang, Yuanda Xu, Zhengze Zhou, Ran He, Zhipeng Wang, and Jiachen Sun. [CRISP: Compressed Reasoning via Iterative Self-Policy Distillation](https://md.research.elamur.ai/free/en/papers/sang-2026-crisp-compressed-reasoning-iterative-self/). arXiv preprint arXiv:2603.05433, 2026. URL https://arxiv.org/abs/2603.05433. [↗](https://arxiv.org/abs/2603.05433)
129. Guobin Shen, Xiang Cheng, Chenxiao Zhao, Lei Huang, Jindong Li, Dongcheng Zhao, and Xing Yu. Anti-self-distillation for reasoning rl via pointwise mutual information. arXiv preprint arXiv:2605.11609v1, 2026a. — [arXiv:2605.11609](https://arxiv.org/abs/2605.11609) [find](https://scholar.google.com/scholar?q=Guobin+Shen%2C+Xiang+Cheng%2C+Chenxiao+Zhao%2C+Lei+Huang%2C+Jindong+Li%2C+Dongcheng+Zhao%2C+and+Xing+Yu%2E+Anti%2Dself%2Ddistillation+for+reasoning+rl+via+pointwise+mutual+information%2E+arXiv+preprint+arXiv%3A2605%2E11609v1)
130. Guobin Shen, Lei Huang, Xiang Cheng, Chenxiao Zhao, Jindong Li, Dongcheng Zhao, and Xing Yu. From generic correlation to input-specific credit in on-policy self distillation. arXiv preprint arXiv:2605.11613v1, 2026b. — [arXiv:2605.11613](https://arxiv.org/abs/2605.11613) [find](https://scholar.google.com/scholar?q=Guobin+Shen%2C+Lei+Huang%2C+Xiang+Cheng%2C+Chenxiao+Zhao%2C+Jindong+Li%2C+Dongcheng+Zhao%2C+and+Xing+Yu%2E+From+generic+correlation+to+input%2Dspecific+credit+in+on%2Dpolicy+self+distillation%2E+arXiv+preprint+arXiv%3A2605)
131. Zhennan Shen, Yanshu Li, Qingyu Yin, Chak Tou Leong, Zhilin Wang, Yanxu Chen, Rongduo Han, Sunbowen Lee, and Yi R. Fung. On the Geometry of On-Policy Distillation. arXiv preprint arXiv:2606.07082, 2026c. URL https://arxiv.org/abs/2606.07082. [↗](https://arxiv.org/abs/2606.07082)
132. Idan Shenfeld, Mehul Damani, Jonas Hübotter, and Pulkit Agrawal. [Self-Distillation Enables Continual Learning](https://md.research.elamur.ai/free/en/papers/shenfeld-2026-self-distillation-enables-continual-learning/). arXiv preprint arXiv:2601.19897, 2026. URL https://arxiv.org/abs/2601.19897. [↗](https://arxiv.org/abs/2601.19897)
133. Guangming Sheng, Chi Cao, Zilingfeng Gao, et al. [HybridFlow: A flexible and efficient RLHF framework](https://md.research.elamur.ai/free/en/papers/sheng-2024-hybridflow-flexible-efficient-rlhf-framework/). arXiv preprint arXiv:2409.19256, 2024.
134. Aasheesh Singh, Vishal Vaddina, and Dagnachew Birru. ORPO-Distill: Mixed-Policy Preference Optimization for Cross-Architecture LLM Distillation. arXiv preprint arXiv:2509.25100, 2025. URL https://arxiv.org/abs/2509.25100. [↗](https://arxiv.org/abs/2509.25100)
135. Yuda Song, Lili Chen, Fahim Tajwar, Remi Munos, Deepak Pathak, J. Andrew Bagnell, Aarti Singh, and Andrea Zanette. Expanding the Capabilities of Reinforcement Learning via Text Feedback. arXiv preprint arXiv:2602.02482, 2026. URL https://arxiv.org/abs/2602.02482. — [arXiv:2602.02482](https://arxiv.org/abs/2602.02482) [↗](https://arxiv.org/abs/2602.02482)
136. Alex Stein, Furong Huang, and Tom Goldstein. GATES: Self-Distillation under Privileged Context with Consensus Gating. arXiv preprint arXiv:2602.20574, 2026. URL https://arxiv.org/abs/2602.20574. — [arXiv:2602.20574](https://arxiv.org/abs/2602.20574) [↗](https://arxiv.org/abs/2602.20574)
137. Xingyu Su, Jacob Helwig, Shubham Parashar, Atharv Chagi, Lakshmi Jotsna, Degui Zhi, James Caverlee, Dileep Kalathil, and Shuiwang Ji. Data-Efficient Autoregressive-to-Diffusion Language Models via On-Policy Distillation. arXiv preprint arXiv:2606.06712, 2026. URL https://arxiv.org/abs/2606.06712. [↗](https://arxiv.org/abs/2606.06712)
138. Jie Sun, Mao Zheng, Mingyang Song, Qiyong Zhong, Yilin Cheng, Bichuan Feng, Pengfei Liu, Junfeng Fang, and Xiang Wang. SimCT: Recovering Lost Supervision for Cross-Tokenizer On-Policy Distillation. arXiv preprint arXiv:2605.07711, 2026. [find](https://scholar.google.com/scholar?q=Jie+Sun%2C+Mao+Zheng%2C+Mingyang+Song%2C+Qiyong+Zhong%2C+Yilin+Cheng%2C+Bichuan+Feng%2C+Pengfei+Liu%2C+Junfeng+Fang%2C+and+Xiang+Wang%2E+SimCT%3A+Recovering+Lost+Supervision+for+Cross%2DTokenizer+On%2DPolicy+Distillation%2E+ar)
139. Wenhui Tan, Minghao Li, Xiaoqian Ma, Siqi Fan, Xiusheng Huang, Liujie Zhang, Ruihua Song, and Weihang Chen. Pair-In, Pair-Out: Latent Multi-Token Prediction for Efficient LLMs. arXiv preprint arXiv:2605.27255, 2026. URL https://arxiv.org/abs/2605.27255. — [arXiv:2605.27255](https://arxiv.org/abs/2605.27255) [↗](https://arxiv.org/abs/2605.27255)
140. Zhiquan Tan and Yinrong Hong. PAINT: Partial-Solution Adaptive Interpolated Training for Self-Distilled Reasoners. arXiv preprint arXiv:2604.26573, 2026a. URL https://arxiv.org/abs/2604.26573. — [arXiv:2604.26573](https://arxiv.org/abs/2604.26573) [↗](https://arxiv.org/abs/2604.26573)
141. Zhiquan Tan and Yinrong Hong. Self-Supervised On-Policy Distillation for Reasoning Language Models. arXiv preprint arXiv:2605.17497, 2026b. URL https://arxiv.org/abs/2605.17497. — [arXiv:2605.17497](https://arxiv.org/abs/2605.17497) [↗](https://arxiv.org/abs/2605.17497)
142. Tencent Robotics X, HY Vision Team, Xumin Yu, Zuyan Liu, Ziyi Wang, He Zhang, Yongming Rao, Fangfu Liu, Yani Zhang, Ruowen Zhao, Oran Wang, Yves Liang, Haitao Lin, Minghui Wang, Yubo Dong, Kevin Cheng, Bolin Ni, Rui Huang, Han Hu, Zhengyou Zhang, Linus, and Shunyu Yao. HY-Embodied-0.5: Embodied Foun… — [arXiv:2604.07430](https://arxiv.org/abs/2604.07430) [find](https://scholar.google.com/scholar?q=Tencent+Robotics+X%2C+HY+Vision+Team%2C+Xumin+Yu%2C+Zuyan+Liu%2C+Ziyi+Wang%2C+He+Zhang%2C+Yongming+Rao%2C+Fangfu+Liu%2C+Yani+Zhang%2C+Ruowen+Zhao%2C+Oran+Wang%2C+Yves+Liang%2C+Haitao+Lin%2C+Minghui+Wang%2C+Yubo+Dong%2C+Kevin+Cheng)
143. Yang Tian, Rui Wang, Xumeng Wen, Junjie Li, Shizhao Sun, Lei Song, Jiang Bian, and Bo Zhao. PBSD: Privileged Bayesian Self-Distillation for Long-Horizon Credit Assignment. arXiv preprint arXiv:2606.09348, 2026. URL https://arxiv.org/abs/2606.09348. — [arXiv:2606.09348](https://arxiv.org/abs/2606.09348) [↗](https://arxiv.org/abs/2606.09348)
144. Bing Wang, Shaotian Yan, Chen Shen, kaiyuan liu, Sinan Fan, Ximing Li, Rui Miao, Xiaosong Yuan, Zhanming Shen, and Jieping Ye. Backtracking When It Strays: Mitigating Dual Exposure Biases in LLM Reasoning Distillation. arXiv preprint arXiv:2605.19433, 2026a. URL https://arxiv.org/abs/2605.19433. — [arXiv:2605.19433](https://arxiv.org/abs/2605.19433) [↗](https://arxiv.org/abs/2605.19433)
145. Hao Wang, Guozhi Wang, Han Xiao, Yufeng Zhou, Yue Pan, Jichao Wang, Ke Xu, Yafei Wen, Xiaohu Ruan, Xiaoxin Chen, and Honggang Qi. Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agents. arXiv preprint arXiv:2604.10674, 2026b. URL https://arxiv.org/abs/2604.10674. — [arXiv:2604.10674](https://arxiv.org/abs/2604.10674) [↗](https://arxiv.org/abs/2604.10674)
146. Jianze Wang, Ying Liu, Jinlong Chen, Xuchun Hu, Qilong Zhang, Yu Cao, Jun Wang, Hua Yang, Yong Xie, and Qianglong Chen. MAD-OPD: Breaking the Ceiling in On-Policy Distillation via Multi-Agent Debate. arXiv preprint arXiv:2605.01347, 2026c. URL https://arxiv.org/abs/2605.01347. — [arXiv:2605.01347](https://arxiv.org/abs/2605.01347) [↗](https://arxiv.org/abs/2605.01347)
147. Jiaqi Wang, Wenhao Zhang, Weijie Shi, Yaliang Li, and James Cheng. TCOD: Exploring Temporal Curriculum in On-Policy Distillation for Multi-turn Autonomous Agents. arXiv preprint arXiv:2604.24005, 2026d. URL https://arxiv.org/abs/2604.24005. — [arXiv:2604.24005](https://arxiv.org/abs/2604.24005) [↗](https://arxiv.org/abs/2604.24005)
148. Jiaxuan Wang, Xuan Ouyang, Zhiyu Chen, Yulan Hu, Zheng Pan, Xin Li, and Lan-Zhe Guo. Trace: Distilling where it matters via token-routed self on-policy alignment. arXiv preprint arXiv:2605.10194v1, 2026e. URL https://arxiv.org/abs/2605.10194. [↗](https://arxiv.org/abs/2605.10194)
149. Ke Wang, Yuning Wu, Haoran Liu, Chaoqun Jia, Devin Chen, and Kai Wei. Physics-Guided Policy Optimization with Self-Distillation. arXiv preprint arXiv:2606.03620, 2026f. URL https://arxiv.org/abs/2606.03620. — [arXiv:2606.03620](https://arxiv.org/abs/2606.03620) [↗](https://arxiv.org/abs/2606.03620)
150. Sudong Wang, Weiquan Huang, Xiaomin Yu, Zuhao Yang, Hehai Lin, Keming Wu, Chaojun Xiao, Chen Chen, Wenxuan Wang, Beier Zhu, Yunjian Zhang, and Chengwei Qin. Beyond SFT-to-RL: Pre-alignment via Black-Box On-Policy Distillation for Multimodal RL. arXiv preprint arXiv:2604.28123, 2026g. URL https://arx… — [arXiv:2604.28123](https://arxiv.org/abs/2604.28123) [find](https://scholar.google.com/scholar?q=Sudong+Wang%2C+Weiquan+Huang%2C+Xiaomin+Yu%2C+Zuhao+Yang%2C+Hehai+Lin%2C+Keming+Wu%2C+Chaojun+Xiao%2C+Chen+Chen%2C+Wenxuan+Wang%2C+Beier+Zhu%2C+Yunjian+Zhang%2C+and+Chengwei+Qin%2E+Beyond+SFT%2Dto%2DRL%3A+Pre%2Dalignment+via+Black%2DB)
151. Wenshuo Wang. Knowledge Distillation Must Account for What It Loses. arXiv preprint arXiv:2604.25110, 2026. — [arXiv:2604.25110](https://arxiv.org/abs/2604.25110) [find](https://scholar.google.com/scholar?q=Wenshuo+Wang%2E+Knowledge+Distillation+Must+Account+for+What+It+Loses%2E+arXiv+preprint+arXiv%3A2604%2E25110%2C+2026%2E+%E2%80%94+%5BarXiv%3A2604%2E25110%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2604%2E25110%29)
152. Xun Wang, Ruishuo Chen, Zhuoran Li, Yu Chen, and Longbo Huang. When Context Returns: Toward Robust Internalization in On-Policy Distillation. arXiv preprint arXiv:2606.11627, 2026h. URL https://arxiv.org/abs/2606.11627. — [arXiv:2606.11627](https://arxiv.org/abs/2606.11627) [↗](https://arxiv.org/abs/2606.11627)
153. Yinjie Wang, Xuyang Chen, Xiaolong Jin, Mengdi Wang, and Ling Yang. OpenClaw-RL: Train Any Agent Simply by Talking. arXiv preprint arXiv:2603.10165, 2026i. URL https://arxiv.org/abs/2603.10165. — [arXiv:2603.10165](https://arxiv.org/abs/2603.10165) [↗](https://arxiv.org/abs/2603.10165)
154. Yuanyi Wang, Su Lu, Yanggan Gu, Pengkai Wang, Yifan Yang, Zhaoyi Yan, Congkai Xie, Jianmin Wu, and Hongxia Yang. Not All Disagreement Is Learnable: Token Teachability in On-Policy Distillation. arXiv preprint arXiv:2605.26844, 2026j. URL https://arxiv.org/abs/2605.26844. — [arXiv:2605.26844](https://arxiv.org/abs/2605.26844) [↗](https://arxiv.org/abs/2605.26844)
155. Zhenlin Wei, Pu Jian, Yingzhuo Deng, Xiaohan Wang, Jiajun Chai, Zhexin Hu, Wei Lin, Shanbin Zhang, and Guojun Yin. AMR-SD: Asymmetric Meta-Reflective Self-Distillation for Token-Level Credit Assignment. arXiv preprint arXiv:2605.18529, 2026. URL https://arxiv.org/abs/2605.18529. — [arXiv:2605.18529](https://arxiv.org/abs/2605.18529) [↗](https://arxiv.org/abs/2605.18529)
156. Ming Wen, Yuxuan Liu, Kun Yang, Yunhao Feng, Zhuoer Xu, Yuhao Sun, Shiwen Cui, Xiang Zheng, Xingjun Ma, and Yu-Gang Jiang. Constitutional On-Policy Safe Distillation. arXiv preprint arXiv:2606.03089, 2026. URL https://arxiv.org/abs/2606.03089. [↗](https://arxiv.org/abs/2606.03089)
157. Yuqiao Wen, Zichao Li, Wenyu Du, and Lili Mou. [f-Divergence Minimization for Sequence-Level Knowledge Distillation](https://md.research.elamur.ai/free/en/papers/wen-2023-f-divergence-minimization-sequence-level/). Proceedings of ACL, 2023. URL https://arxiv.org/abs/2307.15190. [↗](https://arxiv.org/abs/2307.15190)
158. Taiqiang Wu, Chaofan Tao, Jiahao Wang, Runming Yang, Zhe Zhao, and Ngai Wong. [Rethinking Kullback-Leibler Divergence in Knowledge Distillation for Large Language Models](https://md.research.elamur.ai/free/en/papers/wu-2025-rethinking-kullback-leibler-divergence-knowledge/). Proceedings of COLING, 2025. URL https://arxiv.org/abs/2404.02657. [↗](https://arxiv.org/abs/2404.02657)
159. Yecheng Wu, Song Han, and Han Cai. [Lightning OPD: Efficient Post-Training for Large Reasoning Models with Offline On-Policy Distillation](https://md.research.elamur.ai/free/en/papers/wu-2026-lightning-opd-efficient-post-training/). arXiv preprint arXiv:2604.13010, 2026a. URL https://arxiv.org/abs/2604.13010. [↗](https://arxiv.org/abs/2604.13010)
160. Yubin Wu, Zicheng Cai, Liping Ning, Hua Wang, Zhi Chen, Yaohua Tang, and Hao Chen. LiteGUI: Distilling Compact GUI Agents with Reinforcement Learning. arXiv preprint arXiv:2605.07505, 2026b. [find](https://scholar.google.com/scholar?q=Yubin+Wu%2C+Zicheng+Cai%2C+Liping+Ning%2C+Hua+Wang%2C+Zhi+Chen%2C+Yaohua+Tang%2C+and+Hao+Chen%2E+LiteGUI%3A+Distilling+Compact+GUI+Agents+with+Reinforcement+Learning%2E+arXiv+preprint+arXiv%3A2605%2E07505%2C+2026b%2E)
161. Shizhe Xiang, Ke An, Wenlong Yu, Yue Liu, Jian Luan, Pei Fu, and Qilong Wang. Teaching the Way, Not the Answer: Privileged Tutoring Distillation for Multimodal Policy Optimization. arXiv preprint arXiv:2606.07000, 2026. URL https://arxiv.org/abs/2606.07000. — [arXiv:2606.07000](https://arxiv.org/abs/2606.07000) [↗](https://arxiv.org/abs/2606.07000)
162. Xiaomi LLM-Core Team, Bangjun Xiao, Bingquan Xia, Bo Yang, Bofei Gao, Bowen Shen, Chen Zhang, Chenhong He, Chiheng Lou, Fuli Luo, Gang Wang, Gang Xie, Hailin Zhang, Hanglong Lv, Hanyu Li, Heyu Chen, Hongshen Xu, Houbin Zhang, Huaqiu Liu, Jiangshan Duo, Jianyu Wei, Jiebao Xiao, Jinhao Dong, Jun Shi,… — [MiMo-V2-Flash Technical Report](https://md.research.elamur.ai/free/en/papers/team-2026-llm-mimo-v2-flash-technical/)
163. Haoran Xin, Anhao Zhao, Ying Sun, Jin Li, Xiaoyu Shen, and Hui Xiong. Escaping the KL Agreement Trap in On-Policy Distillation. arXiv preprint arXiv:2606.09471, 2026. URL https://arxiv.org/abs/2606.09471. — [arXiv:2606.09471](https://arxiv.org/abs/2606.09471) [↗](https://arxiv.org/abs/2606.09471)
164. Xingrun Xing, Haoqing Wang, Boyan Gao, Ziheng Li, and Yehui Tang. Trust Region On-Policy Distillation. arXiv preprint arXiv:2606.01249, 2026. URL https://arxiv.org/abs/2606.01249. — [arXiv:2606.01249](https://arxiv.org/abs/2606.01249) [↗](https://arxiv.org/abs/2606.01249)
165. Jing Xiong, Hui Shen, Shansan Gong, Yuxin Cheng, Jianghan Shen, Chaofan Tao, Haochen Tan, Haoli Bai, Lifeng Shang, and Ngai Wong. OVD: On-policy Verbal Distillation. arXiv preprint arXiv:2601.21968, 2026. URL https://arxiv.org/abs/2601.21968. — [arXiv:2601.21968](https://arxiv.org/abs/2601.21968) [↗](https://arxiv.org/abs/2601.21968)
166. Haoran Xu, Hongyu Wang, Yifei Gao, Jiaze Li, Xiaofeng Zhang, and Xiaosong Yuan. SG-OPD: Sign-Gated On-Policy Distillation via Sign-Consistency Gating and Phased Teacher Sampling. arXiv preprint arXiv:2606.09304, 2026a. URL https://arxiv.org/abs/2606.09304. — [arXiv:2606.09304](https://arxiv.org/abs/2606.09304) [↗](https://arxiv.org/abs/2606.09304)
167. Hongling Xu, Qi Zhu, Heyuan Deng, Jinpeng Li, Lu Hou, Yasheng Wang, Lifeng Shang, Ruifeng Xu, and Fei Mi. KDRL: Post-Training Reasoning LLMs via Unified Knowledge Distillation and Reinforcement Learning. arXiv preprint arXiv:2506.02208, 2025a. URL https://arxiv.org/abs/2506.02208. — [arXiv:2506.02208](https://arxiv.org/abs/2506.02208) [↗](https://arxiv.org/abs/2506.02208)
168. Shicheng Xu, Liang Pang, Yunchang Zhu, Jia Gu, Zihao Wei, Jingcheng Deng, Feiyang Pan, Huawei Shen, and Xueqi Cheng. RLKD: Distilling LLMs’ Reasoning via Reinforcement Learning. arXiv preprint arXiv:2505.16142, 2025b. URL https://arxiv.org/abs/2505.16142. — [arXiv:2505.16142](https://arxiv.org/abs/2505.16142) [↗](https://arxiv.org/abs/2505.16142)
169. Wenda Xu, Rujun Han, Zifeng Wang, Long T. Le, Dhruv Madeka, Lei Li, William Yang Wang, Rishabh Agarwal, Chen-Yu Lee, and Tomas Pfister. [Speculative Knowledge Distillation: Bridging the Teacher-Student Gap Through Interleaved Sampling](https://md.research.elamur.ai/free/en/papers/xu-2024-speculative-knowledge-distillation-bridging-teacher/). Proceedings of ICLR, 2025c. URL https://arxiv.org/abs/2410.11325. [↗](https://arxiv.org/abs/2410.11325)
170. Xiaohan Xu, Ming Li, Chongyang Tao, Tao Shen, Reynold Cheng, Jinyang Li, Can Xu, Dacheng Tao, and Tianyi Zhou. A Survey on Knowledge Distillation of Large Language Models. arXiv preprint arXiv:2402.13116, 2024. URL https://arxiv.org/abs/2402.13116. — [arXiv:2402.13116](https://arxiv.org/abs/2402.13116) [↗](https://arxiv.org/abs/2402.13116)
171. Yuanda Xu, Hejian Sang, Zhengze Zhou, Ran He, and Zhipeng Wang. PACED: Distillation and On-Policy Self-Distillation at the Frontier of Student Competence. arXiv preprint arXiv:2603.11178, 2026b. URL https://arxiv.org/abs/2603.11178. — [arXiv:2603.11178](https://arxiv.org/abs/2603.11178) [↗](https://arxiv.org/abs/2603.11178)
172. Yuanda Xu, Hejian Sang, Zhengze Zhou, Ran He, Zhipeng Wang, and Alborz Geramifard. [TIP: Token Importance in On-Policy Distillation](https://md.research.elamur.ai/free/en/papers/xu-2026-tip-token-importance-policy-distillation/). arXiv preprint arXiv:2604.14084, 2026c. URL https://arxiv.org/abs/2604.14084. [↗](https://arxiv.org/abs/2604.14084)
173. Yuanda Xu, Hejian Sang, Zhengze Zhou, Ran He, Zhipeng Wang, and Alborz Geramifard. Beyond grpo and on-policy distillation: An empirical sparse-to-dense reward principle for language-model post-training. arXiv preprint arXiv:2605.12483v1, 2026d. [find](https://scholar.google.com/scholar?q=Yuanda+Xu%2C+Hejian+Sang%2C+Zhengze+Zhou%2C+Ran+He%2C+Zhipeng+Wang%2C+and+Alborz+Geramifard%2E+Beyond+grpo+and+on%2Dpolicy+distillation%3A+An+empirical+sparse%2Dto%2Ddense+reward+principle+for+language%2Dmodel+post%2Dtrainin)
174. Jianhao Yan, Yafu Li, Zican Hu, Zhi Wang, Ganqu Cui, Xiaoye Qu, Yu Cheng, and Yue Zhang. [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, 2025. URL https://arxiv.org/abs/2504.14945. [↗](https://arxiv.org/abs/2504.14945)
175. Shaotian Yan, Kaiyuan Liu, Chen Shen, Bing Wang, Sinan Fan, Jun Zhang, Yue Wu, Zheng Wang, and Jieping Ye. Distribution-Aligned Sequence Distillation for Superior Long-CoT Reasoning. arXiv preprint arXiv:2601.09088, 2026. URL https://arxiv.org/abs/2601.09088. [↗](https://arxiv.org/abs/2601.09088)
176. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, Chujie Zheng, Dayiheng Liu, Fan Zhou, Fei Huang, Feng Hu, Hao Ge, Haoran Wei, Huan Lin, Jialong Tang, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jing Zhou, Jin… — [Qwen3 Technical Report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/)
177. Chenxu Yang, Chuanyu Qin, Qingyi Si, Minghui Chen, Naibin Gu, Dingyu Yao, Zheng Lin, Weiping Wang, Jiaqi Wang, and Nan Duan. [Self-Distilled RLVR](https://md.research.elamur.ai/free/en/papers/yang-2026-self-distilled-rlvr/). arXiv preprint arXiv:2604.03128, 2026a. URL https://arxiv.org/abs/2604.03128. [↗](https://arxiv.org/abs/2604.03128)
178. Fan Yang, Rui Meng, Trudi Di Qi, Ali Ezzati, and Yuxin Wen. KEPO: Knowledge-Enhanced Preference Optimization for Multimodal Reasoning with Applications to Medical VQA. arXiv preprint arXiv:2602.00400, 2026b. URL https://arxiv.org/abs/2602.00400. — [arXiv:2602.00400](https://arxiv.org/abs/2602.00400) [↗](https://arxiv.org/abs/2602.00400)
179. Han Yang, Mingyan Wu, Bailan He, Zeyu Cao, Sikuan Yan, Kevin Qinghong Lin, and Zifeng Ding. Reasoning compression with mixed-policy distillation. arXiv preprint arXiv:2605.08776, 2026c. — [arXiv:2605.08776](https://arxiv.org/abs/2605.08776) [find](https://scholar.google.com/scholar?q=Han+Yang%2C+Mingyan+Wu%2C+Bailan+He%2C+Zeyu+Cao%2C+Sikuan+Yan%2C+Kevin+Qinghong+Lin%2C+and+Zifeng+Ding%2E+Reasoning+compression+with+mixed%2Dpolicy+distillation%2E+arXiv+preprint+arXiv%3A2605%2E08776%2C+2026c%2E+%E2%80%94+%5BarXiv%3A2605%2E)
180. Ling Yang, Zhaochen Yu, Tianjun Zhang, Minkai Xu, Joseph E. Gonzalez, Bin Cui, and Shuicheng Yan. SuperCorrect: Advancing Small LLM Reasoning with Thought Template Distillation and Self-Correction. Proceedings of ICLR, 2025b. URL https://arxiv.org/abs/2410.09008. — [arXiv:2410.09008](https://arxiv.org/abs/2410.09008) [↗](https://arxiv.org/abs/2410.09008)
181. Shenzhi Yang, Guangcheng Zhu, Bowen Song, Haobo Wang, Mingxuan Xia, Xing Zheng, Yingfan Ma, Zhongqi Chen, Weiqiang Wang, and Gang Chen. OPRD: On-Policy Representation Distillation. arXiv preprint arXiv:2606.06021, 2026d. URL https://arxiv.org/abs/2606.06021. — [arXiv:2606.06021](https://arxiv.org/abs/2606.06021) [↗](https://arxiv.org/abs/2606.06021)
182. Wenkai Yang, Weijie Liu, Ruobing Xie, Kai Yang, Saiyong Yang, and Yankai Lin. Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation. arXiv preprint arXiv:2602.12125, 2026e. URL https://arxiv.org/abs/2602.12125. — [arXiv:2602.12125](https://arxiv.org/abs/2602.12125) [↗](https://arxiv.org/abs/2602.12125)
183. Yuxiao Yang, Xiaoyun Wang, and Weitong Zhang. Ogls-sd: On-policy self-distillation with outcome-guided logit steering for llm reasoning. arXiv preprint arXiv:2605.12400v1, 2026f. — [arXiv:2605.12400](https://arxiv.org/abs/2605.12400) [find](https://scholar.google.com/scholar?q=Yuxiao+Yang%2C+Xiaoyun+Wang%2C+and+Weitong+Zhang%2E+Ogls%2Dsd%3A+On%2Dpolicy+self%2Ddistillation+with+outcome%2Dguided+logit+steering+for+llm+reasoning%2E+arXiv+preprint+arXiv%3A2605%2E12400v1%2C+2026f%2E+%E2%80%94+%5BarXiv%3A2605%2E12400%5D%28)
184. Zhicheng Yang, Zhijiang Guo, Yifan Song, Minrui Xu, Yongxin Wang, Yiwei Wang, Xiaodan Liang, and Jing Tang. Prune-OPD: Efficient and Reliable On-Policy Distillation for Long-Horizon Reasoning. arXiv preprint arXiv:2605.07804, 2026g. — [arXiv:2605.07804](https://arxiv.org/abs/2605.07804) [find](https://scholar.google.com/scholar?q=Zhicheng+Yang%2C+Zhijiang+Guo%2C+Yifan+Song%2C+Minrui+Xu%2C+Yongxin+Wang%2C+Yiwei+Wang%2C+Xiaodan+Liang%2C+and+Jing+Tang%2E+Prune%2DOPD%3A+Efficient+and+Reliable+On%2DPolicy+Distillation+for+Long%2DHorizon+Reasoning%2E+arXiv+p)
185. Zhuolin Yang, Zihan Liu, Yang Chen, Wenliang Dai, Boxin Wang, Sheng-Chieh Lin, Chankyu Lee, Yangyi Chen, Dongfu Jiang, Jiafan He, Renjie Pi, Grace Lam, Nayeon Lee, Alexander Bukharin, Mohammad Shoeybi, Bryan Catanzaro, and Wei Ping. Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Do… — [Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation](https://md.research.elamur.ai/free/en/papers/yang-2026-nemotron-cascade-2-post-training/)
186. Tianzhu Ye, Li Dong, Zewen Chi, Xun Wu, Shaohan Huang, and Furu Wei. Black-Box On-Policy Distillation of Large Language Models. arXiv preprint arXiv:2511.10643, 2025. URL https://arxiv.org/abs/2511.10643. — [arXiv:2511.10643](https://arxiv.org/abs/2511.10643) [↗](https://arxiv.org/abs/2511.10643)
187. Tianzhu Ye, Li Dong, Qingxiu Dong, Xun Wu, Shaohan Huang, and Furu Wei. Online Experiential Learning for Language Models. arXiv preprint arXiv:2603.16856, 2026a. URL https://arxiv.org/abs/2603.16856. — [arXiv:2603.16856](https://arxiv.org/abs/2603.16856) [↗](https://arxiv.org/abs/2603.16856)
188. Tianzhu Ye, Li Dong, Xun Wu, Shaohan Huang, and Furu Wei. [On-Policy Context Distillation for Language Models](https://md.research.elamur.ai/free/en/papers/ye-2026-policy-context-distillation-language-models/). arXiv preprint arXiv:2602.12275, 2026b. URL https://arxiv.org/abs/2602.12275. [↗](https://arxiv.org/abs/2602.12275)
189. Woongyeng Yeo, Yumin Choi, Taekyung Ki, and Sung Ju Hwang. HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents. arXiv preprint arXiv:2605.17873, 2026. URL https://arxiv.org/abs/2605.17873. — [arXiv:2605.17873](https://arxiv.org/abs/2605.17873) [↗](https://arxiv.org/abs/2605.17873)
190. Hee Suk Yoon, Eunseop Yoon, Jaehyun Jang, SooHwan Eom, Ji Woo Hong, Mark Hasegawa-Johnson, Qi Dai, Chong Luo, and Chang D. Yoo. Decomposed On-Policy Distillation for Vision-Language Reasoning: Steering Gradients for Visual Grounding. arXiv preprint arXiv:2606.00564, 2026. URL https://arxiv.org/abs/2… — [arXiv:2606.00564](https://arxiv.org/abs/2606.00564) [↗](https://arxiv.org/abs/2…)
191. Guo Yu, Wenlin Liu, Yulan Hu, Hao-Xuan Ma, Jun-Peng Jiang, and Han-Jia Ye. Dense Supervision, Sparse Updates: On the Sparsity and Geometry of On-Policy Distillation. arXiv preprint arXiv:2606.13657, 2026a. URL https://arxiv.org/abs/2606.13657. — [arXiv:2606.13657](https://arxiv.org/abs/2606.13657) [↗](https://arxiv.org/abs/2606.13657)
192. Weichen Yu, Xiaomin Li, Yizhou Zhao, Xiaoze Liu, Ruowang Zhang, Haixin Wang, Yinyi Luo, Chen Henry Wu, Gaurav Mittal, Matt Fredrikson, and Yu Hu. Multi-rollout on-policy distillation via peer successes and failures. arXiv preprint arXiv:2605.12652, 2026b. — [arXiv:2605.12652](https://arxiv.org/abs/2605.12652) [find](https://scholar.google.com/scholar?q=Weichen+Yu%2C+Xiaomin+Li%2C+Yizhou+Zhao%2C+Xiaoze+Liu%2C+Ruowang+Zhang%2C+Haixin+Wang%2C+Yinyi+Luo%2C+Chen+Henry+Wu%2C+Gaurav+Mittal%2C+Matt+Fredrikson%2C+and+Yu+Hu%2E+Multi%2Drollout+on%2Dpolicy+distillation+via+peer+successe)
193. Xin Yu, Liuchen Liao, Yiwen Zhang, Yingchen Yu, Lingzhou Xue, and Qinzhen Guo. Preference-based self-distillation: Beyond KL matching via reward regularization. arXiv preprint arXiv:2605.05040, 2026c. [find](https://scholar.google.com/scholar?q=Xin+Yu%2C+Liuchen+Liao%2C+Yiwen+Zhang%2C+Yingchen+Yu%2C+Lingzhou+Xue%2C+and+Qinzhen+Guo%2E+Preference%2Dbased+self%2Ddistillation%3A+Beyond+KL+matching+via+reward+regularization%2E+arXiv+preprint+arXiv%3A2605%2E05040%2C+2026c%2E)
194. Qianhao Yuan, Jie Lou, Xing Yu, Hongyu Lin, Le Sun, Xianpei Han, and Yaojie Lu. Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation. arXiv preprint arXiv:2605.18740, 2026. URL https://arxiv.org/abs/2605.18740. — [arXiv:2605.18740](https://arxiv.org/abs/2605.18740) [↗](https://arxiv.org/abs/2605.18740)
195. Azure Zhang, Ning Gao, Yuqin Dai, Ruiyuan Wu, Jinpeng Wang, Rena Wei Gao, Bingdong Tan, Shuzheng Gao, Zongjie Li, and Chaozheng Wang. Unlocking Proactivity in Task-Oriented Dialogue. arXiv preprint arXiv:2605.22240, 2026a. URL https://arxiv.org/abs/2605.22240. — [arXiv:2605.22240](https://arxiv.org/abs/2605.22240) [↗](https://arxiv.org/abs/2605.22240)
196. Dongxu Zhang, Zhichao Yang, Sepehr Janghorbani, Jun Han, Andrew Ressler, Qian Qian, Gregory D. Lyng, Sanjit Singh Batra, and Robert E. Tillman. [Fast and Effective On-policy Distillation from Reasoning Prefixes](https://md.research.elamur.ai/free/en/papers/zhang-2026-fast-effective-policy-distillation-reasoning/). arXiv preprint arXiv:2602.15260, 2026b. URL https://arxiv.org/abs/2602.15260. [↗](https://arxiv.org/abs/2602.15260)
197. Hongbin Zhang, Chaozheng Wang, Kehai Chen, Youcheng Pan, Yang Xiang, Jinpeng Wang, and Min Zhang. Tailoring Teaching to Aptitude: Direction-Adaptive Self-Distillation for LLM Reasoning. arXiv preprint arXiv:2605.22263, 2026c. URL https://arxiv.org/abs/2605.22263. [↗](https://arxiv.org/abs/2605.22263)
198. Jiaxin Zhang, Xiangyu Peng, Qinglin Chen, Qinyuan Ye, Caiming Xiong, and Chien-Sheng Wu. The Illusion of Certainty: Decoupling Capability and Calibration in On-Policy Distillation. arXiv preprint arXiv:2604.16830, 2026d. URL https://arxiv.org/abs/2604.16830. [↗](https://arxiv.org/abs/2604.16830)
199. Ke Zhang, Yunjie Tian, DongDi Zhao, Yijiang Li, Yuanye Liu, Vishal M Patel, and Di Fu. On-policy distillation with best-of-n teacher rollout selection. arXiv preprint arXiv:2605.09725, 2026e. — [arXiv:2605.09725](https://arxiv.org/abs/2605.09725) [find](https://scholar.google.com/scholar?q=Ke+Zhang%2C+Yunjie+Tian%2C+DongDi+Zhao%2C+Yijiang+Li%2C+Yuanye+Liu%2C+Vishal+M+Patel%2C+and+Di+Fu%2E+On%2Dpolicy+distillation+with+best%2Dof%2Dn+teacher+rollout+selection%2E+arXiv+preprint+arXiv%3A2605%2E09725%2C+2026e%2E+%E2%80%94+%5BarXiv)
200. Miaosen Zhang, Yishan Liu, Shuxia Lin, Xu Yang, Qi Dai, Chong Luo, Weihao Jiang, Peng Hou, Anxiang Zeng, Xin Geng, and Baining Guo. Towards On-Policy SFT: Distribution Discriminant Theory and its Applications in LLM Training. arXiv preprint arXiv:2602.12222, 2026f. URL https://arxiv.org/abs/2602.122… — [arXiv:2602.12222](https://arxiv.org/abs/2602.12222) [↗](https://arxiv.org/abs/2602.122…)
201. Songming Zhang, Xue Zhang, Tong Zhang, Bojie Hu, Yufeng Chen, and Jinan Xu. AlignDistil: Token-Level Language Model Alignment as Adaptive Policy Distillation. In Proceedings of ACL, 2025a. URL https://arxiv.org/abs/2503.02832. [↗](https://arxiv.org/abs/2503.02832)
202. Wenhao Zhang. Beyond Absolute Imitation: Anchored Residual Guidance for Privileged On-Policy Distillation. arXiv preprint arXiv:2606.10385, 2026. URL https://arxiv.org/abs/2606.10385. — [arXiv:2606.10385](https://arxiv.org/abs/2606.10385) [↗](https://arxiv.org/abs/2606.10385)
203. Xinsen Zhang, Zhenkai Ding, Tianjun Pan, Run Yang, Chun Kang, Xue Xiong, and Jingnan Gu. OPSDL: On-Policy Self-Distillation for Long-Context Language Models. arXiv preprint arXiv:2604.17535, 2026g. URL https://arxiv.org/abs/2604.17535. — [arXiv:2604.17535](https://arxiv.org/abs/2604.17535) [↗](https://arxiv.org/abs/2604.17535)
204. XiuYu Zhang, Yi Shan, Junfeng Fang, and Zhenkai Liang. Self-Evaluation Is Already There: Eliciting Latent Judge Calibration in Base LLMs with Minimal Data. arXiv preprint arXiv:2606.05122, 2026h. URL https://arxiv.org/abs/2606.05122. — [arXiv:2606.05122](https://arxiv.org/abs/2606.05122) [↗](https://arxiv.org/abs/2606.05122)
205. Xue Zhang, Songming Zhang, Yunlong Liang, Fandong Meng, Yufeng Chen, Jinan Xu, and Jie Zhou. A Dual-Space Framework for General Knowledge Distillation of Large Language Models. arXiv preprint arXiv:2504.11426, 2025b. URL https://arxiv.org/abs/2504.11426. — [arXiv:2504.11426](https://arxiv.org/abs/2504.11426) [↗](https://arxiv.org/abs/2504.11426)
206. Yan Zhang, Daiqing Wu, Huawen Shen, Yu Zhou, and Can Ma. Learn where to click from yourself: On-policy self-distillation for gui grounding. arXiv preprint arXiv:2605.00642, 2026i. — [arXiv:2605.00642](https://arxiv.org/abs/2605.00642) [find](https://scholar.google.com/scholar?q=Yan+Zhang%2C+Daiqing+Wu%2C+Huawen+Shen%2C+Yu+Zhou%2C+and+Can+Ma%2E+Learn+where+to+click+from+yourself%3A+On%2Dpolicy+self%2Ddistillation+for+gui+grounding%2E+arXiv+preprint+arXiv%3A2605%2E00642%2C+2026i%2E+%E2%80%94+%5BarXiv%3A2605%2E00642%5D)
207. Yanfei Zhang, Xu Lin, and Chenglin Wu. StepOPSD: Step-Aware Online Preference Distillation for Agent Reinforcement Learning. arXiv preprint arXiv:2605.27140, 2026j. URL https://arxiv.org/abs/2605.27140. — [arXiv:2605.27140](https://arxiv.org/abs/2605.27140) [↗](https://arxiv.org/abs/2605.27140)
208. Yaocheng Zhang, Jiajun Chai, Yuqian Fu, Songjun Tu, Xiaohan Wang, Wei Lin, Guojun Yin, Qichao Zhang, Yuanheng Zhu, and Dongbin Zhao. Are Full Rollouts Necessary for On-Policy Distillation? arXiv preprint arXiv:2605.31490, 2026k. URL https://arxiv.org/abs/2605.31490. — [arXiv:2605.31490](https://arxiv.org/abs/2605.31490) [↗](https://arxiv.org/abs/2605.31490)
209. Yaocheng Zhang, Yuanheng Zhu, Wenyue Chong, Songjun Tu, Qichao Zhang, Jiajun Chai, Xiaohan Wang, Wei Lin, Guojun Yin, and Dongbin Zhao. π-Play: Multi-Agent Self-Play via Privileged Self-Distillation without External Data. arXiv preprint arXiv:2604.14054, 2026l. URL https://arxiv.org/abs/2604.14054. — [arXiv:2604.14054](https://arxiv.org/abs/2604.14054) [↗](https://arxiv.org/abs/2604.14054)
210. Yulin Zhang, He Cao, Zihao Jiang, Chenyi Zi, Zhipeng Zhou, Zijing Liu, Yu Li, Jia Li, and Ziqi Gao. Proteinopd: Towards effective and efficient preference alignment for protein design. arXiv preprint arXiv:2605.10189v1, 2026m. URL https://arxiv.org/abs/2605.10189. — [arXiv:2605.10189](https://arxiv.org/abs/2605.10189) [↗](https://arxiv.org/abs/2605.10189)
211. Yuwei Zhang, Sha Li, Changlong Yu, Qin Lu, Shuowei Jin, Chengyu Dong, Haoran Liu, Ilgee Hong, Xintong Li, Zhenyu Shi, Bing Yin, and Jingbo Shang. Learning with rare success but rich feedback via reflection-enhanced self-distillation. arXiv preprint arXiv:2605.12741, 2026n. — [arXiv:2605.12741](https://arxiv.org/abs/2605.12741) [find](https://scholar.google.com/scholar?q=Yuwei+Zhang%2C+Sha+Li%2C+Changlong+Yu%2C+Qin+Lu%2C+Shuowei+Jin%2C+Chengyu+Dong%2C+Haoran+Liu%2C+Ilgee+Hong%2C+Xintong+Li%2C+Zhenyu+Shi%2C+Bing+Yin%2C+and+Jingbo+Shang%2E+Learning+with+rare+success+but+rich+feedback+via+refle)
212. Zhaoyang Zhang, Shuli Jiang, Yantao Shen, Yuting Zhang, Dhananjay Ram, Shuo Yang, Zhuowen Tu, Wei Xia, and Stefano Soatto. [Reinforcement-aware Knowledge Distillation for LLM Reasoning](https://md.research.elamur.ai/free/en/papers/zhang-2026-reinforcement-aware-knowledge-distillation-llm/). arXiv preprint arXiv:2602.22495, 2026o. URL https://arxiv.org/abs/2602.22495. [↗](https://arxiv.org/abs/2602.22495)
213. Hanyang Zhao, Haoxian Chen, Han Lin, Genta Indra Winata, David Yao, and Wenpin Tang. OPD+: Rethinking the Advantage Design for On-Policy Distillation. arXiv preprint arXiv:2606.01039, 2026a. URL https://arxiv.org/abs/2606.01039. — [arXiv:2606.01039](https://arxiv.org/abs/2606.01039) [↗](https://arxiv.org/abs/2606.01039)
214. Siyan Zhao, Zhihui Xie, Mengchen Liu, Jing Huang, Guan Pang, Feiyu Chen, and Aditya Grover. [Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models](https://md.research.elamur.ai/free/en/papers/zhao-2026-self-distilled-reasoner-policy-self/). arXiv preprint arXiv:2601.18734, 2026b. URL https://arxiv.org/abs/2601.18734. [↗](https://arxiv.org/abs/2601.18734)
215. Zhengyang Zhao, Lu Ma, and Wentao Zhang. Training with harnesses: On-policy harness self-distillation for complex reasoning. arXiv preprint arXiv:2605.08741v1, 2026c. — [arXiv:2605.08741](https://arxiv.org/abs/2605.08741) [find](https://scholar.google.com/scholar?q=Zhengyang+Zhao%2C+Lu+Ma%2C+and+Wentao+Zhang%2E+Training+with+harnesses%3A+On%2Dpolicy+harness+self%2Ddistillation+for+complex+reasoning%2E+arXiv+preprint+arXiv%3A2605%2E08741v1%2C+2026c%2E+%E2%80%94+%5BarXiv%3A2605%2E08741%5D%28https%3A%2F%2Farxi)
216. Ziqi Zhao, Xinyu Ma, Liu Yang, Yujie Feng, Daiting Shi, Jingzhou He, Xin Xin, Zhaochun Ren, and Xiao-Ming Wu. ROSD: Reflective On-Policy Self-Distillation for Language Model Reasoning across Domains. arXiv preprint arXiv:2605.28014, 2026d. URL https://arxiv.org/abs/2605.28014. [↗](https://arxiv.org/abs/2605.28014)
217. Binbin Zheng, Xing Ma, Yiheng Liang, Jingqing Ruan, Xiaoliang Fu, Kepeng Lin, Benchang Zhu, Ke Zeng, and Xunliang Cai. SCOPE: Signal-Calibrated On-Policy Distillation Enhancement with Dual-Path Adaptive Weighting. arXiv preprint arXiv:2604.10688, 2026a. URL https://arxiv.org/abs/2604.10688. — [arXiv:2604.10688](https://arxiv.org/abs/2604.10688) [↗](https://arxiv.org/abs/2604.10688)
218. Haoyu Zheng, Yun Zhu, Shu Yuan, Shangming Chen, Qing Wang, Wenqiao Zhang, Jun Xiao, and Yueting Zhuang. MAIGO: Mitigating Lost-in-Conversation with History-Cleaned On-Policy Self-Distillation. arXiv preprint arXiv:2605.27186, 2026b. URL https://arxiv.org/abs/2605.27186. — [arXiv:2605.27186](https://arxiv.org/abs/2605.27186) [↗](https://arxiv.org/abs/2605.27186)
219. Qihuang Zhong, Liang Ding, Li Shen, Juhua Liu, Bo Du, and Dacheng Tao. Revisiting Knowledge Distillation for Autoregressive Language Models. Proceedings of ACL, 2024. URL https://arxiv.org/abs/2402.11890. [↗](https://arxiv.org/abs/2402.11890)
220. Qiyong Zhong, Mao Zheng, Mingyang Song, Xin Lin, Jie Sun, Houcheng Jiang, Xiang Wang, and Junfeng Fang. SOD: Step-wise On-policy Distillation for Small Language Model Agents. arXiv preprint arXiv:2605.07725, 2026a. — [arXiv:2605.07725](https://arxiv.org/abs/2605.07725) [find](https://scholar.google.com/scholar?q=Qiyong+Zhong%2C+Mao+Zheng%2C+Mingyang+Song%2C+Xin+Lin%2C+Jie+Sun%2C+Houcheng+Jiang%2C+Xiang+Wang%2C+and+Junfeng+Fang%2E+SOD%3A+Step%2Dwise+On%2Dpolicy+Distillation+for+Small+Language+Model+Agents%2E+arXiv+preprint+arXiv%3A2605)
221. Zhide Zhong, Haodong Yan, Junfeng Li, Junjie He, Tianran Zhang, and Haoang Li. VLA-OPD: Bridging Offline SFT and Online RL for Vision-Language-Action Models via On-Policy Distillation. arXiv preprint arXiv:2603.26666, 2026b. URL https://arxiv.org/abs/2603.26666. — [arXiv:2603.26666](https://arxiv.org/abs/2603.26666) [↗](https://arxiv.org/abs/2603.26666)
222. Yongchao Zhou, Kaifeng Lyu, Ankit Singh Rawat, Aditya Krishna Menon, Afshin Rostamizadeh, Sanjiv Kumar, Jean-François Kagy, and Rishabh Agarwal. DistillSpec: Improving Speculative Decoding via Knowledge Distillation. Proceedings of ICLR, 2024. URL https://arxiv.org/abs/2310.08461. — [arXiv:2310.08461](https://arxiv.org/abs/2310.08461) [↗](https://arxiv.org/abs/2310.08461)
223. Yucheng Zhou, Wei Tao, Yiwen Guo, and Jianbing Shen. World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning. arXiv preprint arXiv:2606.03603, 2026a. URL https://arxiv.org/abs/2606.03603. — [arXiv:2606.03603](https://arxiv.org/abs/2606.03603) [↗](https://arxiv.org/abs/2606.03603)
224. Yuhang Zhou, Lizhu Zhang, Yifan Wu, Mingyi Wang, Peng Bo, Jiayi Liu, Xiangjun Fan, and Zhuokai Zhao. OmniOPD: Logit-Free On-Policy Distillation via Speculative Verification. arXiv preprint arXiv:2606.01476, 2026b. URL https://arxiv.org/abs/2606.01476. — [arXiv:2606.01476](https://arxiv.org/abs/2606.01476) [↗](https://arxiv.org/abs/2606.01476)
225. Siqi Zhu, Xuyan Ye, Hongyu Lu, Weiye Shi, and Ge Liu. The many faces of on-policy distillation: Pitfalls, mechanisms, and fixes. arXiv preprint arXiv:2605.11182v1, 2026. — [arXiv:2605.11182](https://arxiv.org/abs/2605.11182) [find](https://scholar.google.com/scholar?q=Siqi+Zhu%2C+Xuyan+Ye%2C+Hongyu+Lu%2C+Weiye+Shi%2C+and+Ge+Liu%2E+The+many+faces+of+on%2Dpolicy+distillation%3A+Pitfalls%2C+mechanisms%2C+and+fixes%2E+arXiv+preprint+arXiv%3A2605%2E11182v1%2C+2026%2E+%E2%80%94+%5BarXiv%3A2605%2E11182%5D%28https%3A%2F%2Fa)
226. Zhou Ziheng, Jiaqi Li, Huacong Tang, Ying Nian Wu, and Demetri Terzopoulos. Less is More: Early Stopping Rollout for On-Policy Distillation. arXiv preprint arXiv:2605.27028, 2026. URL https://arxiv.org/abs/2605.27028. [↗](https://arxiv.org/abs/2605.27028)
227. Matthieu Zimmer, Xiaotong Ji, Tu Nguyen, and Haitham Bou Ammar. Rethinking Large Language Model Distillation: A Constrained Markov Decision Process Perspective. arXiv preprint arXiv:2509.22921, 2025. URL https://arxiv.org/abs/2509.22921. — [arXiv:2509.22921](https://arxiv.org/abs/2509.22921) [↗](https://arxiv.org/abs/2509.22921)
228. Qingyun Zou, Yifan Zhang, Bin Lin, and Wei Chen. Reward-weighted on-policy distillation with an open property-equivalence verifier for NL-to-SVA generation. arXiv preprint arXiv:2605.13501, 2026. — [arXiv:2605.13501](https://arxiv.org/abs/2605.13501) [find](https://scholar.google.com/scholar?q=Qingyun+Zou%2C+Yifan+Zhang%2C+Bin+Lin%2C+and+Wei+Chen%2E+Reward%2Dweighted+on%2Dpolicy+distillation+with+an+open+property%2Dequivalence+verifier+for+NL%2Dto%2DSVA+generation%2E+arXiv+preprint+arXiv%3A2605%2E13501%2C+2026%2E+%E2%80%94+%5Ba)

*Updated 3 November 2026*