---
title: "CODERL+ — paper by Jiang et al. (2026) — summary and ideas"
description: "LLMs learn code by predicting text, but whether a program is correct depends on how it behaves when… 56 references, cited by 4 papers here."
authors: ["Xue Jiang", "Yihong Dong", "Mengyang Liu", "Deng Hongyi", "Tian Wang", "Yongding Tao", "Zhi Jin", "Wenpin Jiao", "Ge Li"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2026.acl-long.164/"
doi: "10.18653/v1/2026.acl-long.164"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/jiang-2026-coderl-improving-code-generation-reinforcement/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/jiang-2026-coderl-improving-code-generation-reinforcement/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/3/) / CODERL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment

# CODERL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment

**Authors:** [Xue Jiang](https://md.research.elamur.ai/free/en/authors/xue-jiang/), [Yihong Dong](https://md.research.elamur.ai/free/en/authors/yihong-dong/), Mengyang Liu, Deng Hongyi, [Tian Wang](https://md.research.elamur.ai/free/en/authors/tian-wang/), [Yongding Tao](https://md.research.elamur.ai/free/en/authors/yongding-tao/), [Zhi Jin](https://md.research.elamur.ai/free/en/authors/zhi-jin/), Wenpin Jiao and [Ge Li](https://md.research.elamur.ai/free/en/authors/ge-li/) · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.acl-long.164/) · [PDF](https://aclanthology.org/2026.acl-long.164.pdf) · [DOI 10.18653/v1/2026.acl-long.164](https://doi.org/10.18653/v1/2026.acl-long.164) · [OpenAlex](https://openalex.org/W4415967387)

**Citations:** 1 citations (OpenAlex, 2026-10-14) · [4](https://md.research.elamur.ai/free/en/papers/jiang-2026-coderl-improving-code-generation-reinforcement/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary

LLMs learn code by predicting text, but whether a program is correct depends on how it behaves when run. RLVR methods try to close this gap by rewarding programs that pass tests. A single pass/fail bit, however, gives little guidance about subtle logic errors such as mistracked loop variables.

CODERL+ adds a second, parallel RL objective to standard RLVR training. The model must infer the variable-level execution trajectory of programs, meaning the final values variables take while the code runs. These tasks are built on the fly from the policy's own failed code-generation rollouts, so no extra dataset is needed and the tasks adapt as the model improves. The method works with different RL algorithms, with GRPO as the main base.

Compared with RLVR and distillation baselines, CODERL+ improves pass@1 on benchmarks including HumanEval, LeetCode and LiveCodeBench. It also transfers to code reasoning and test-output prediction. This matters because it shows that dense supervision from execution states gives more useful training signal than outcome-only rewards.

## Key points

- Average relative pass@1 gain of 4.6% over post-training baselines (RLVR, Distillation). Accuracy rises by 15.5% on code-reasoning and 4.4% on test-output-generation benchmarks.
- In a motivating trace-inference experiment, RLVR improved only about 4% over the base model. The paper's figure reports an 81.8% relative improvement for CODERL+ on that task.
- Ablations support the design. Random rather than failed rollouts gave training samples that were too easy. Pre-built (off-policy) alignment data performed worse. Supervising only input/output (IO) instead of intermediate variable states lowered results on all tasks.
- Linear regression probes on hidden states at variable tokens predicted runtime values with lower MSE for CODERL+ than for the base or GRPO models at every layer. The gap was largest in the middle layers.
- Limitations: experiments cover only models up to 8B parameters, no hyperparameter tuning was done (each run takes about 3 days), and training costs extra compute. The code is released at github.com/jiangxxxue/CODERLPLUS.

*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

- [Peking University](https://md.research.elamur.ai/free/en/universities/peking-university/) — University, Beijing, China · 244 works · 39,359 citations

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 1,437 works

🧭 Topics (1): [💻 Code & programming languages](https://md.research.elamur.ai/free/en/topics/code-and-programming/) · 129 works

## Cite

Cite as: Xue Jiang, Yihong Dong, Mengyang Liu, Deng Hongyi, Tian Wang, Yongding Tao, Zhi Jin, Wenpin Jiao and Ge Li. 2026. CODERL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

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

```bibtex
@inproceedings{jiang2026coderl,
  title = {{CODERL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment}},
  author = {Xue Jiang, Yihong Dong, Mengyang Liu, Deng Hongyi, Tian Wang, Yongding Tao, Zhi Jin, Wenpin Jiao and Ge Li},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.164},
  url = {https://doi.org/10.18653/v1/2026.acl-long.164}
}
```

- [← previous ChatR1: Reinforcement Learning for Conversational Reasoning and Retrieval Augmented Question Answering](https://md.research.elamur.ai/free/en/papers/lupart-2026-chatr1-reinforcement-learning-conversational-reasoning/)
- [next → Coding Agents with Multimodal Browsing are Generalist Problem Solvers](https://md.research.elamur.ai/free/en/papers/soni-2026-coding-agents-multimodal-browsing-are/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/3/)

## References

56 references, 20 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Wasi Uddin Ahmad, Sean Narenthiran, Somshubra Majumdar, Aleksander Ficek, Siddhartha Jain, Jocelyn Huang, Vahid Noroozi, and Boris Ginsburg. 2025. [Opencodereasoning: Advancing data distillation for competitive coding](https://md.research.elamur.ai/free/en/papers/ahmad-2025-opencodereasoning-advancing-data-distillation-competitive/). arXiv preprint arXiv:2504.01943.
2. Jacob Austin, Augustus Odena, Maxwell I. Nye, Maarten Bosma, Henryk Michalewski, David Dohan, Ellen Jiang, Carrie J. Cai, Michael Terry, Quoc V. Le, and Charles Sutton. 2021. [Program synthesis with large language models](https://md.research.elamur.ai/free/en/papers/austin-2021-program-synthesis-large-language-models/). CoRR, abs/2108.07732.
3. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Pondé de Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, Alex Ray, Raul Puri, Gretchen Krueger, Michael Petrov, Heidy Khlaaf, Girish Sastry, Pamela Mishkin, Brooke Chan, Scott Gray, and 39 others. 202… — [Evaluating Large Language Models Trained on Code](https://md.research.elamur.ai/free/en/papers/chen-2021-evaluating-large-language-models-trained/)
4. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit S. Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, Luke Marris, Sam Petulla, Colin Gaffney, Asaf Aharoni, Nathan Lintz, Tiago Cardal Pais, Henrik Jacobsson, Idan Szpektor, Nan-Jiang Jiang, and 81 othe… — [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://md.research.elamur.ai/free/en/papers/comanici-2025-gemini-2-5-pushing-frontier/)
5. Ganqu Cui, Lifan Yuan, Zefan Wang, Hanbin Wang, Wendi Li, Bingxiang He, Yuchen Fan, Tianyu Yu, Qixin Xu, Weize Chen, Jiarui Yuan, Huayu Chen, Kaiyan Zhang, Xingtai Lv, Shuo Wang, Yuan Yao, Xu Han, Hao Peng, Yu Cheng, and 4 others. 2025. [Process reinforcement through implicit rewards](https://md.research.elamur.ai/free/en/papers/cui-2025-process-reinforcement-through-implicit-rewards/). arXiv preprint…
6. Yangruibo Ding, Jinjun Peng, Marcus Min, Gail Kaiser, Junfeng Yang, and Baishakhi Ray. 2024. Semcoder: Training code language models with comprehensive semantics reasoning. Advances in Neural Information Processing Systems (NeurIPS), pages 60275– 60308. [find](https://scholar.google.com/scholar?q=Yangruibo+Ding%2C+Jinjun+Peng%2C+Marcus+Min%2C+Gail+Kaiser%2C+Junfeng+Yang%2C+and+Baishakhi+Ray%2E+2024%2E+Semcoder%3A+Training+code+language+models+with+comprehensive+semantics+reasoning%2E+Advances+in+Neural+Informat)
7. Yihong Dong, Jiazheng Ding, Xue Jiang, Ge Li, Zhuo Li, and Zhi Jin. 2023. Codescore: Evaluating code generation by learning code execution. CoRR, abs/2301.09043. — [arXiv:2301.09043](https://arxiv.org/abs/2301.09043) [find](https://scholar.google.com/scholar?q=Yihong+Dong%2C+Jiazheng+Ding%2C+Xue+Jiang%2C+Ge+Li%2C+Zhuo+Li%2C+and+Zhi+Jin%2E+2023%2E+Codescore%3A+Evaluating+code+generation+by+learning+code+execution%2E+CoRR%2C+abs%2F2301%2E09043%2E+%E2%80%94+%5BarXiv%3A2301%2E09043%5D%28https%3A%2F%2Farxiv%2Eorg)
8. Yihong Dong, Xue Jiang, Zhi Jin, and Ge Li. 2024a. Self-collaboration code generation via chatgpt. ACM Trans. Softw. Eng. Methodol., 33(7):189:1–189:38. [find](https://scholar.google.com/scholar?q=Yihong+Dong%2C+Xue+Jiang%2C+Zhi+Jin%2C+and+Ge+Li%2E+2024a%2E+Self%2Dcollaboration+code+generation+via+chatgpt%2E+ACM+Trans%2E+Softw%2E+Eng%2E+Methodol%2E%2C+33%287%29%3A189%3A1%E2%80%93189%3A38%2E)
9. Yihong Dong, Xue Jiang, Huanyu Liu, Zhi Jin, Bin Gu, Mengfei Yang, and Ge Li. 2024b. Generalization or memorization: Data contamination and trustworthy evaluation for large language models. In ACL (Findings), pages 12039–12050. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Yihong+Dong%2C+Xue+Jiang%2C+Huanyu+Liu%2C+Zhi+Jin%2C+Bin+Gu%2C+Mengfei+Yang%2C+and+Ge+Li%2E+2024b%2E+Generalization+or+memorization%3A+Data+contamination+and+trustworthy+evaluation+for+large+language+models%2E+In+ACL+%28Fi)
10. Yihong Dong, Xue Jiang, Jiaru Qian, Tian Wang, Kechi Zhang, Zhi Jin, and Ge Li. 2025a. A survey on code generation with llm-based agents. CoRR, abs/2508.00083. — [arXiv:2508.00083](https://arxiv.org/abs/2508.00083) [find](https://scholar.google.com/scholar?q=Yihong+Dong%2C+Xue+Jiang%2C+Jiaru+Qian%2C+Tian+Wang%2C+Kechi+Zhang%2C+Zhi+Jin%2C+and+Ge+Li%2E+2025a%2E+A+survey+on+code+generation+with+llm%2Dbased+agents%2E+CoRR%2C+abs%2F2508%2E00083%2E+%E2%80%94+%5BarXiv%3A2508%2E00083%5D%28https%3A%2F%2Farxiv%2Eorg%2Fa)
11. Yihong Dong, Xue Jiang, Yongding Tao, Huanyu Liu, Kechi Zhang, Lili Mou, Rongyu Cao, Yingwei Ma, Jue Chen, Binhua Li, Zhi Jin, Fei Huang, Yongbin Li, and Ge Li. 2025b. [RL-PLUS: countering capability boundary collapse of llms in reinforcement learning with hybrid-policy optimization](https://md.research.elamur.ai/free/en/papers/dong-2026-rl-plus-countering-capability-boundary/). CoRR, abs/2508.0…
12. Shihan Dou, Yan Liu, Haoxiang Jia, Enyu Zhou, Limao Xiong, Junjie Shan, Caishuang Huang, Xiao Wang, Xiaoran Fan, Zhiheng Xi, Yuhao Zhou, Tao Ji, Rui Zheng, Qi Zhang, Tao Gui, and Xuanjing Huang. 2024. Stepcoder: Improving code generation with reinforcement learning from compiler feedback. In Annual… [find](https://scholar.google.com/scholar?q=Shihan+Dou%2C+Yan+Liu%2C+Haoxiang+Jia%2C+Enyu+Zhou%2C+Limao+Xiong%2C+Junjie+Shan%2C+Caishuang+Huang%2C+Xiao+Wang%2C+Xiaoran+Fan%2C+Zhiheng+Xi%2C+Yuhao+Zhou%2C+Tao+Ji%2C+Rui+Zheng%2C+Qi+Zhang%2C+Tao+Gui%2C+and+Xuanjing+Huang%2E+2024%2E)
13. Xueying Du, Mingwei Liu, Kaixin Wang, Hanlin Wang, Junwei Liu, Yixuan Chen, Jiayi Feng, Chaofeng Sha, Xin Peng, and Yiling Lou. 2024. Evaluating large language models in class-level code generation. In ACM. [find](https://scholar.google.com/scholar?q=Xueying+Du%2C+Mingwei+Liu%2C+Kaixin+Wang%2C+Hanlin+Wang%2C+Junwei+Liu%2C+Yixuan+Chen%2C+Jiayi+Feng%2C+Chaofeng+Sha%2C+Xin+Peng%2C+and+Yiling+Lou%2E+2024%2E+Evaluating+large+language+models+in+class%2Dlevel+code+generation%2E+I)
14. Meta FAIR CodeGen Team. 2025. CWM: An open-weights LLM for research on code generation with world models. [find](https://scholar.google.com/scholar?q=Meta+FAIR+CodeGen+Team%2E+2025%2E+CWM%3A+An+open%2Dweights+LLM+for+research+on+code+generation+with+world+models%2E)
15. Lishui Fan, Yu Zhang, Mouxiang Chen, and Zhongxin Liu. 2025. Posterior-grpo: Rewarding reasoning processes in code generation. arXiv preprint arXiv:2508.05170. — [arXiv:2508.05170](https://arxiv.org/abs/2508.05170) [find](https://scholar.google.com/scholar?q=Lishui+Fan%2C+Yu+Zhang%2C+Mouxiang+Chen%2C+and+Zhongxin+Liu%2E+2025%2E+Posterior%2Dgrpo%3A+Rewarding+reasoning+processes+in+code+generation%2E+arXiv+preprint+arXiv%3A2508%2E05170%2E+%E2%80%94+%5BarXiv%3A2508%2E05170%5D%28https%3A%2F%2Farxiv%2Eorg%2Fa)
16. Yuxian Gu, Li Dong, Furu Wei, and Minlie Huang. 2023. [Minillm: Knowledge distillation of large language models](https://md.research.elamur.ai/free/en/papers/gu-2023-minillm-policy-distillation-large-language/). arXiv preprint arXiv:2306.08543.
17. Arnav Gudibande, Eric Wallace, Charlie Snell, Xinyang Geng, Hao Liu, Pieter Abbeel, Sergey Levine, and Dawn Song. 2023. [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.
18. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, and 80 others. 2025. DeepSeek-R1 incentivizes reasoning in LLMs through reinf… [find](https://scholar.google.com/scholar?q=Daya+Guo%2C+Dejian+Yang%2C+Haowei+Zhang%2C+Junxiao+Song%2C+Ruoyu+Zhang%2C+Runxin+Xu%2C+Qihao+Zhu%2C+Shirong+Ma%2C+Peiyi+Wang%2C+Xiao+Bi%2C+Xiaokang+Zhang%2C+Xingkai+Yu%2C+Yu+Wu%2C+Z%2E+F%2E+Wu%2C+Zhibin+Gou%2C+Zhihong+Shao%2C+Zhuoshu+Li)
19. Alex Havrilla, Yuqing Du, Sharath Chandra Raparthy, Christoforos Nalmpantis, Jane Dwivedi-Yu, Maksym Zhuravinskyi, Eric Hambro, Sainbayar Sukhbaatar, and Roberta Raileanu. 2024. Teaching large language models to reason with reinforcement learning. arXiv preprint arXiv:2403.04642. [find](https://scholar.google.com/scholar?q=Alex+Havrilla%2C+Yuqing+Du%2C+Sharath+Chandra+Raparthy%2C+Christoforos+Nalmpantis%2C+Jane+Dwivedi%2DYu%2C+Maksym+Zhuravinskyi%2C+Eric+Hambro%2C+Sainbayar+Sukhbaatar%2C+and+Roberta+Raileanu%2E+2024%2E+Teaching+large+languag)
20. Jujie He, Jiacai Liu, Chris Yuhao Liu, Rui Yan, Chaojie Wang, Peng Cheng, Xiaoyu Zhang, Fuxiang Zhang, Jiacheng Xu, Wei Shen, Siyuan Li, Liang Zeng, Tianwen Wei, Cheng Cheng, Bo An, Yang Liu, and Yahui Zhou. 2025. Skywork open reasoner 1 technical report. arXiv preprint arXiv:2505.22312. [find](https://scholar.google.com/scholar?q=Jujie+He%2C+Jiacai+Liu%2C+Chris+Yuhao+Liu%2C+Rui+Yan%2C+Chaojie+Wang%2C+Peng+Cheng%2C+Xiaoyu+Zhang%2C+Fuxiang+Zhang%2C+Jiacheng+Xu%2C+Wei+Shen%2C+Siyuan+Li%2C+Liang+Zeng%2C+Tianwen+Wei%2C+Cheng+Cheng%2C+Bo+An%2C+Yang+Liu%2C+and+Yahu)
21. Dan Hendrycks, Steven Basart, Saurav Kadavath, Mantas Mazeika, Akul Arora, Ethan Guo, Collin Burns, Samir Puranik, Horace He, Dawn Song, and Jacob Steinhardt. 2021. [Measuring coding challenge competence with apps](https://md.research.elamur.ai/free/en/papers/hendrycks-2021-measuring-coding-challenge-competence-apps/). arXiv preprint arXiv:2105.09938.
22. John Hewitt and Percy Liang. 2019. Designing and interpreting probes with control tasks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 2733–2743. [find](https://scholar.google.com/scholar?q=John+Hewitt+and+Percy+Liang%2E+2019%2E+Designing+and+interpreting+probes+with+control+tasks%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Methods+in+Natural+Language+Processing+and+the+9th+Internati)
23. Hugging Face. 2025. Open r1: A fully open reproduction of deepseek-r1. [find](https://scholar.google.com/scholar?q=Hugging+Face%2E+2025%2E+Open+r1%3A+A+fully+open+reproduction+of+deepseek%2Dr1%2E)
24. Binyuan Hui, Jian Yang, Zeyu Cui, Jiaxi Yang, Dayiheng Liu, Lei Zhang, Tianyu Liu, Jiajun Zhang, Bowen Yu, Kai Dang, An Yang, Rui Men, Fei Huang, Xingzhang Ren, Xuancheng Ren, Jingren Zhou, and Junyang Lin. 2024. Qwen2. 5-coder technical report. arXiv preprint arXiv:2409.12186. — [Qwen2.5-Coder Technical Report](https://md.research.elamur.ai/free/en/papers/hui-2024-qwen2-5-coder-technical-report/)
25. Naman Jain, King Han, Alex Gu, Wen-Ding Li, Fanjia Yan, Tianjun Zhang, Sida Wang, Armando Solar-Lezama, Koushik Sen, and Ion Stoica. 2024. [Livecodebench: Holistic and contamination free evaluation of large language models for code](https://md.research.elamur.ai/free/en/papers/jain-2024-livecodebench-holistic-contamination-free-evaluation/). arXiv preprint arXiv:2403.07974.
26. Xue Jiang, Yihong Dong, Zhi Jin, and Ge Li. 2024a. SEED: customize large language models with sample-efficient adaptation for code generation. CoRR, abs/2403.00046. — [arXiv:2403.00046](https://arxiv.org/abs/2403.00046) [find](https://scholar.google.com/scholar?q=Xue+Jiang%2C+Yihong+Dong%2C+Zhi+Jin%2C+and+Ge+Li%2E+2024a%2E+SEED%3A+customize+large+language+models+with+sample%2Defficient+adaptation+for+code+generation%2E+CoRR%2C+abs%2F2403%2E00046%2E+%E2%80%94+%5BarXiv%3A2403%2E00046%5D%28https%3A%2F%2Farxiv%2E)
27. Xue Jiang, Yihong Dong, Yongding Tao, Huanyu Liu, Zhi Jin, and Ge Li. 2025. ROCODE: integrating backtracking mechanism and program analysis in large language models for code generation. In ICSE, pages 334–346. IEEE. [find](https://scholar.google.com/scholar?q=Xue+Jiang%2C+Yihong+Dong%2C+Yongding+Tao%2C+Huanyu+Liu%2C+Zhi+Jin%2C+and+Ge+Li%2E+2025%2E+ROCODE%3A+integrating+backtracking+mechanism+and+program+analysis+in+large+language+models+for+code+generation%2E+In+ICSE%2C+pages)
28. Xue Jiang, Yihong Dong, Lecheng Wang, Zheng Fang, Qiwei Shang, Ge Li, Zhi Jin, and Wenpin Jiao. 2024b. Self-planning code generation with large language models. ACM Trans. Softw. Eng. Methodol., 33(7):182:1–182:30. [find](https://scholar.google.com/scholar?q=Xue+Jiang%2C+Yihong+Dong%2C+Lecheng+Wang%2C+Zheng+Fang%2C+Qiwei+Shang%2C+Ge+Li%2C+Zhi+Jin%2C+and+Wenpin+Jiao%2E+2024b%2E+Self%2Dplanning+code+generation+with+large+language+models%2E+ACM+Trans%2E+Softw%2E+Eng%2E+Methodol%2E%2C+33%287%29)
29. Tianjie Ju, Weiwei Sun, Wei Du, Xinwei Yuan, Zhaochun Ren, and Gongshen Liu. 2024. How large language models encode context knowledge? a layer-wise probing study. arXiv preprint arXiv:2402.16061. [find](https://scholar.google.com/scholar?q=Tianjie+Ju%2C+Weiwei+Sun%2C+Wei+Du%2C+Xinwei+Yuan%2C+Zhaochun+Ren%2C+and+Gongshen+Liu%2E+2024%2E+How+large+language+models+encode+context+knowledge%3F+a+layer%2Dwise+probing+study%2E+arXiv+preprint+arXiv%3A2402%2E16061%2E)
30. Hung Le, Yue Wang, Akhilesh Deepak Gotmare, Silvio Savarese, and Steven Chu Hong Hoi. 2022. [CodeRL: Mastering code generation through pretrained models and deep reinforcement learning](https://md.research.elamur.ai/free/en/papers/le-2022-coderl-mastering-code-generation-through/). Advances in Neural Information Processing Systems (NeurIPS), pages 21314–21328.
31. Cuong Le Chi, Chau Truong Vinh Hoang, Phan Nhat Huy, Dung D Le, Tien N Nguyen, and Nghi DQ Bui. 2025. VisualCoder: Guiding large language models in code execution with fine-grained multimodal chain-of-thought reasoning. In Findings of the Annual Conference of the Nations of the Americas Chapter of t… [find](https://scholar.google.com/scholar?q=Cuong+Le+Chi%2C+Chau+Truong+Vinh+Hoang%2C+Phan+Nhat+Huy%2C+Dung+D+Le%2C+Tien+N+Nguyen%2C+and+Nghi+DQ+Bui%2E+2025%2E+VisualCoder%3A+Guiding+large+language+models+in+code+execution+with+fine%2Dgrained+multimodal+chain%2Dof)
32. Jason D Lee, Qi Lei, Nikunj Saunshi, and Jiacheng Zhuo. 2021. Predicting what you already know helps: Provable self-supervised learning. Advances in Neural Information Processing Systems, 34:309–323. [find](https://scholar.google.com/scholar?q=Jason+D+Lee%2C+Qi+Lei%2C+Nikunj+Saunshi%2C+and+Jiacheng+Zhuo%2E+2021%2E+Predicting+what+you+already+know+helps%3A+Provable+self%2Dsupervised+learning%2E+Advances+in+Neural+Information+Processing+Systems%2C+34%3A309%E2%80%93323%2E)
33. Junlong Li, Daya Guo, Dejian Yang, Runxin Xu, Yu Wu, and Junxian He. 2025a. CodeIO: Condensing reasoning patterns via code input-output prediction. In International Conference on Machine Learning (ICML). [find](https://scholar.google.com/scholar?q=Junlong+Li%2C+Daya+Guo%2C+Dejian+Yang%2C+Runxin+Xu%2C+Yu+Wu%2C+and+Junxian+He%2E+2025a%2E+CodeIO%3A+Condensing+reasoning+patterns+via+code+input%2Doutput+prediction%2E+In+International+Conference+on+Machine+Learning+%28ICM)
34. Qingyao Li, Xinyi Dai, Xiangyang Li, Weinan Zhang, Yasheng Wang, Ruiming Tang, and Yong Yu. 2025b. CodePRM: Execution feedback-enhanced process reward model for code generation. In Findings of the Annual Meeting of the Association for Computational Linguistics (ACL), pages 8169–8182. [find](https://scholar.google.com/scholar?q=Qingyao+Li%2C+Xinyi+Dai%2C+Xiangyang+Li%2C+Weinan+Zhang%2C+Yasheng+Wang%2C+Ruiming+Tang%2C+and+Yong+Yu%2E+2025b%2E+CodePRM%3A+Execution+feedback%2Denhanced+process+reward+model+for+code+generation%2E+In+Findings+of+the+Ann)
35. Rongao Li, Jie Fu, Bo-Wen Zhang, Tao Huang, Zhihong Sun, Chen Lyu, Guang Liu, Zhi Jin, and Ge Li. 2023. Taco: Topics in algorithmic code generation dataset. arXiv preprint arXiv:2312.14852. [find](https://scholar.google.com/scholar?q=Rongao+Li%2C+Jie+Fu%2C+Bo%2DWen+Zhang%2C+Tao+Huang%2C+Zhihong+Sun%2C+Chen+Lyu%2C+Guang+Liu%2C+Zhi+Jin%2C+and+Ge+Li%2E+2023%2E+Taco%3A+Topics+in+algorithmic+code+generation+dataset%2E+arXiv+preprint+arXiv%3A2312%2E14852%2E)
36. Yujia Li, David Choi, Junyoung Chung, Nate Kushman, Julian Schrittwieser, Rémi Leblond, Tom Eccles, James Keeling, Felix Gimeno, Agustin Dal Lago, Thomas Hubert, Peter Choy, Cyprien de Masson d’Autume, Igor Babuschkin, Xinyun Chen, Po-Sen Huang, Johannes Welbl, Sven Gowal, Alexey Cherepanov, and 7 o… — [Competition-Level Code Generation with AlphaCode](https://md.research.elamur.ai/free/en/papers/li-2022-competition-level-code-generation-alphacode/)
37. Jiawei Liu, Chunqiu Steven Xia, Yuyao Wang, and Lingming Zhang. 2023. [Is your code generated by chatgpt really correct? rigorous evaluation of large language models for code generation](https://md.research.elamur.ai/free/en/papers/liu-2023-your-code-generated-chatgpt-really/). In Advances in Neural Information Processing Systems 36, pages 21558– 21572.
38. Meta AI. 2024. Introducing llama 3.1: Our most capable models to date. [https://ai.meta.com/blog/](https://ai.meta.com/blog/) meta-llama-3-1/. Accessed: 2025-10-06. [↗](https://ai.meta.com/blog/)
39. OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, and 261 others.… — [GPT-4 Technical Report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/)
40. Guilherme Penedo, Anton Lozhkov, Hynek Ky-dlíček, Loubna Ben Allal, Edward Beeching, Agustín Piqueres Lajarín, Quentin Gallouédec, Nathan Habib, Lewis Tunstall, and Leandro von Werra. 2025. Codeforces. [https://huggingface.co/datasets/open-r1/codeforces](https://huggingface.co/datasets/open-r1/codeforces). [↗](https://huggingface.co/datasets/open-r1/codeforces)
41. David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R Bowman. 2024. [Gpqa: A graduate-level google-proof q&a benchmark](https://md.research.elamur.ai/free/en/papers/rein-2023-gpqa-graduate-level-google-proof/). In First Conference on Language Modeling.
42. Baptiste Rozière, Jonas Gehring, Fabian Gloeckle, Sten Sootla, Itai Gat, Xiaoqing Ellen Tan, Yossi Adi, Jingyu Liu, Romain Sauvestre, Tal Remez, Jérémy Rapin, Artyom Kozhevnikov, Ivan Evtimov, Joanna Bitton, Manish Bhatt, Cristian Canton Ferrer, Aaron Grattafiori, Wenhan Xiong, Alexandre Défossez, a… — [Code Llama: Open Foundation Models for Code](https://md.research.elamur.ai/free/en/papers/roziere-2024-code-llama-open-foundation-models/)
43. John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. 2017. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347. [find](https://scholar.google.com/scholar?q=John+Schulman%2C+Filip+Wolski%2C+Prafulla+Dhariwal%2C+Alec+Radford%2C+and+Oleg+Klimov%2E+2017%2E+Proximal+policy+optimization+algorithms%2E+arXiv+preprint+arXiv%3A1707%2E06347%2E)
44. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Mingchuan Zhang, Y. K. Li, Y. Wu, and Daya Guo. 2024. [Deepseekmath: Pushing the limits of mathematical reasoning in open language models](https://md.research.elamur.ai/free/en/papers/shao-2024-deepseekmath-pushing-limits-mathematical-reasoning/). CoRR, abs/2402.03300.
45. Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, and Chuan Wu. 2024. [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.
46. Lingxiao Tang, He Ye, Zhongxin Liu, Xiaoxue Ren, and Lingfeng Bao. 2025. Codereasoner: Enhancing the code reasoning ability with reinforcement learning. arXiv preprint arXiv:2507.17548. — [arXiv:2507.17548](https://arxiv.org/abs/2507.17548) [find](https://scholar.google.com/scholar?q=Lingxiao+Tang%2C+He+Ye%2C+Zhongxin+Liu%2C+Xiaoxue+Ren%2C+and+Lingfeng+Bao%2E+2025%2E+Codereasoner%3A+Enhancing+the+code+reasoning+ability+with+reinforcement+learning%2E+arXiv+preprint+arXiv%3A2507%2E17548%2E+%E2%80%94+%5BarXiv%3A2507%2E)
47. Miles Turpin, Julian Michael, Ethan Perez, and Samuel Bowman. 2023. [Language models don’t always say what they think: Unfaithful explanations in chain-of-thought prompting](https://md.research.elamur.ai/free/en/papers/turpin-2023-language-models-don-t-always/). Advances in Neural Information Processing Systems, 36:74952–74965.
48. Sijie Wang, Quanjiang Guo, Kai Zhao, Yawei Zhang, Xin Li, Xiang Li, Siqi Li, Rui She, Shangshu Yu, and Wee Peng Tay. 2025a. CodeBoost: Boosting code LLMs by squeezing knowledge from code snippets with rl. arXiv preprint arXiv:2508.05242. — [arXiv:2508.05242](https://arxiv.org/abs/2508.05242) [find](https://scholar.google.com/scholar?q=Sijie+Wang%2C+Quanjiang+Guo%2C+Kai+Zhao%2C+Yawei+Zhang%2C+Xin+Li%2C+Xiang+Li%2C+Siqi+Li%2C+Rui+She%2C+Shangshu+Yu%2C+and+Wee+Peng+Tay%2E+2025a%2E+CodeBoost%3A+Boosting+code+LLMs+by+squeezing+knowledge+from+code+snippets+with)
49. Yanlin Wang, Tianyue Jiang, Mingwei Liu, Jiachi Chen, Mingzhi Mao, Xilin Liu, Yuchi Ma, and Zibin Zheng. 2025b. Beyond functional correctness: Investigating coding style inconsistencies in large language models. Proceedings of the ACM on Software Engineering, 2(FSE):690–712. [find](https://scholar.google.com/scholar?q=Yanlin+Wang%2C+Tianyue+Jiang%2C+Mingwei+Liu%2C+Jiachi+Chen%2C+Mingzhi+Mao%2C+Xilin+Liu%2C+Yuchi+Ma%2C+and+Zibin+Zheng%2E+2025b%2E+Beyond+functional+correctness%3A+Investigating+coding+style+inconsistencies+in+large+langu)
50. Yanlin Wang, Yanli Wang, Daya Guo, Jiachi Chen, Ruikai Zhang, Yuchi Ma, and Zibin Zheng. 2024. RLCoder: Reinforcement learning for repository-level code completion. In International Conference on Software Engineering (ICSE), pages 165–177. [find](https://scholar.google.com/scholar?q=Yanlin+Wang%2C+Yanli+Wang%2C+Daya+Guo%2C+Jiachi+Chen%2C+Ruikai+Zhang%2C+Yuchi+Ma%2C+and+Zibin+Zheng%2E+2024%2E+RLCoder%3A+Reinforcement+learning+for+repository%2Dlevel+code+completion%2E+In+International+Conference+on+Soft)
51. Yinjie Wang, Ling Yang, Ye Tian, Ke Shen, and Mengdi Wang. 2025c. Co-evolving LLM coder and unit tester via reinforcement learning. CoRR, abs/2506.03136. — [arXiv:2506.03136](https://arxiv.org/abs/2506.03136) [find](https://scholar.google.com/scholar?q=Yinjie+Wang%2C+Ling+Yang%2C+Ye+Tian%2C+Ke+Shen%2C+and+Mengdi+Wang%2E+2025c%2E+Co%2Devolving+LLM+coder+and+unit+tester+via+reinforcement+learning%2E+CoRR%2C+abs%2F2506%2E03136%2E+%E2%80%94+%5BarXiv%3A2506%2E03136%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F250)
52. Colin White, Samuel Dooley, Manley Roberts, Arka Pal, Benjamin Feuer, Siddhartha Jain, Ravid Shwartz-Ziv, Neel Jain, Khalid Saifullah, Sreemanti Dey, Shubh-Agrawal, Sandeep Singh Sandha, Siddartha V. Naidu, Chinmay Hegde, Yann LeCun, Tom Goldstein, Willie Neiswanger, and Micah Goldblum. 2025. Livebe… [find](https://scholar.google.com/scholar?q=Colin+White%2C+Samuel+Dooley%2C+Manley+Roberts%2C+Arka+Pal%2C+Benjamin+Feuer%2C+Siddhartha+Jain%2C+Ravid+Shwartz%2DZiv%2C+Neel+Jain%2C+Khalid+Saifullah%2C+Sreemanti+Dey%2C+Shubh%2DAgrawal%2C+Sandeep+Singh+Sandha%2C+Siddartha+V%2E+)
53. Haoze Wu, Yunzhi Yao, Wenhao Yu, Huajun Chen, and Ningyu Zhang. 2025. ReCode: Updating code api knowledge with reinforcement learning. arXiv preprint arXiv:2506.20495. — [arXiv:2506.20495](https://arxiv.org/abs/2506.20495) [find](https://scholar.google.com/scholar?q=Haoze+Wu%2C+Yunzhi+Yao%2C+Wenhao+Yu%2C+Huajun+Chen%2C+and+Ningyu+Zhang%2E+2025%2E+ReCode%3A+Updating+code+api+knowledge+with+reinforcement+learning%2E+arXiv+preprint+arXiv%3A2506%2E20495%2E+%E2%80%94+%5BarXiv%3A2506%2E20495%5D%28https%3A%2F%2Farx)
54. Yunhui Xia, Wei Shen, Yan Wang, Jason Klein Liu, Huifeng Sun, Siyue Wu, Jian Hu, and Xiaolong Xu. 2025. Leetcodedataset: A temporal dataset for robust evaluation and efficient training of code llms. arXiv preprint arXiv:2504.14655. — [arXiv:2504.14655](https://arxiv.org/abs/2504.14655) [find](https://scholar.google.com/scholar?q=Yunhui+Xia%2C+Wei+Shen%2C+Yan+Wang%2C+Jason+Klein+Liu%2C+Huifeng+Sun%2C+Siyue+Wu%2C+Jian+Hu%2C+and+Xiaolong+Xu%2E+2025%2E+Leetcodedataset%3A+A+temporal+dataset+for+robust+evaluation+and+efficient+training+of+code+llms%2E+a)
55. Xiaohan Xu, Ming Li, Chongyang Tao, Tao Shen, Reynold Cheng, Jinyang Li, Can Xu, Dacheng Tao, and Tianyi Zhou. 2024. A survey on knowledge distillation of large language models. arXiv preprint arXiv:2402.13116. — [arXiv:2402.13116](https://arxiv.org/abs/2402.13116) [find](https://scholar.google.com/scholar?q=Xiaohan+Xu%2C+Ming+Li%2C+Chongyang+Tao%2C+Tao+Shen%2C+Reynold+Cheng%2C+Jinyang+Li%2C+Can+Xu%2C+Dacheng+Tao%2C+and+Tianyi+Zhou%2E+2024%2E+A+survey+on+knowledge+distillation+of+large+language+models%2E+arXiv+preprint+arXiv%3A2)
56. Hao Yu, Bo Shen, Dezhi Ran, Jiaxin Zhang, Qi Zhang, Yuchi Ma, Guangtai Liang, Ying Li, Qianxiang Wang, and Tao Xie. 2024. Codereval: A benchmark of pragmatic code generation with generative pre-trained models. In ACM. [find](https://scholar.google.com/scholar?q=Hao+Yu%2C+Bo+Shen%2C+Dezhi+Ran%2C+Jiaxin+Zhang%2C+Qi+Zhang%2C+Yuchi+Ma%2C+Guangtai+Liang%2C+Ying+Li%2C+Qianxiang+Wang%2C+and+Tao+Xie%2E+2024%2E+Codereval%3A+A+benchmark+of+pragmatic+code+generation+with+generative+pre%2Dtraine)

*Updated 31 October 2026*