---
title: "UCoder — paper by Wu et al. (2026) — summary and ideas"
description: "Both are costly to build at scale. 6.7 points. 34 references, cited by 1 paper here. Among them: 0."
authors: ["Jiajun Wu", "Jian Yang", "Wei Zhang", "Linzheng Chai", "Yuchi Ma", "Ensheng Shi", "Yuqing Ma", "Zhoujun Li", "Xianglong Liu"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.277/"
doi: "10.18653/v1/2026.findings-acl.277"
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/wu-2026-ucoder-unsupervised-code-generation-internal/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wu-2026-ucoder-unsupervised-code-generation-internal/)

[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/18/) / UCoder: Unsupervised Code Generation by Internal Probing of Large Language Models

# UCoder: Unsupervised Code Generation by Internal Probing of Large Language Models

**Authors:** [Jiajun Wu](https://md.research.elamur.ai/free/en/authors/jiajun-wu/), [Jian Yang](https://md.research.elamur.ai/free/en/authors/jian-yang/), [Wei Zhang](https://md.research.elamur.ai/free/en/authors/wei-zhang/), [Linzheng Chai](https://md.research.elamur.ai/free/en/authors/linzheng-chai/), Yuchi Ma, Ensheng Shi, Yuqing Ma, [Zhoujun Li](https://md.research.elamur.ai/free/en/authors/zhoujun-li/) and [Xianglong Liu](https://md.research.elamur.ai/free/en/authors/xianglong-liu/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/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.findings-acl.277/) · [PDF](https://aclanthology.org/2026.findings-acl.277.pdf) · [DOI 10.18653/v1/2026.findings-acl.277](https://doi.org/10.18653/v1/2026.findings-acl.277) · [OpenAlex](https://openalex.org/W7166852194)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · 1 in our library, of the 149,439 papers whose bibliographies we read

## Summary
Code-focused LLMs usually improve through supervised instruction tuning on curated problem–solution pairs, or on synthetic data derived from real code snippets. Both are costly to build at scale. This paper asks whether a model can improve its coding using only knowledge from pre-training, with no external corpus at all.

The authors propose IPC, a six-stage self-bootstrapping pipeline:
1. The base model writes algorithmic problems.
2. It rates their difficulty.
3. It drafts function skeletons.
4. It produces about a hundred test inputs per problem.
5. It densely samples candidate solutions and runs them on those tests. Candidates are grouped by identical outputs, on the idea that correct programs agree while wrong ones fail in different ways. Each candidate is scored by execution success, cluster size (consensus) and perplexity-based fluency.
6. The selected solutions are used for supervised fine-tuning, and the loop repeats with the improved model.

The resulting UCoder models (7B, 14B, 32B) come close to supervised instruct baselines on several code benchmarks. This suggests that internal model signals can replace labeled data when resources are limited.

## Key points
- Evaluation covers HumanEval(+), MBPP(+), BigCodeBench (Complete/Instruct, Full/Hard), LiveCodeBench and FullStackBench. Comparisons include Qwen2.5-Coder, DS-Coder, CodeLlama, GPT-4o and Claude-3.5-Sonnet.
- UCoder-7B reaches 85.2 on MBPP, versus 83.5 for Qwen2.5-Coder-7B-Instruct. At 14B and 32B, UCoder generally trails the instruct models on the BigCodeBench Hard splits.
- Solution space probing samples 128 candidates per problem. Across 9,700 candidates, perplexity averaged about 1.03, and solutions with over 80% execution success mostly fell below 1.05.
- In the ablation, consensus-based selection beat Random, Cluster, Low PPL and Success Rate (0.5 threshold) selection. On FullStackBench its margin over Random grew from 6.7 points at 7B to 13.9 points at 32B.
- Filtering raised success rate, error-free rate and code uniqueness while keeping structural and semantic entropy roughly unchanged.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 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 (3): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works · [💻 Code & programming languages](https://md.research.elamur.ai/free/en/topics/code-and-programming/) · 129 works · [🔬 Interpretability](https://md.research.elamur.ai/free/en/topics/interpretability/) · 130 works

## Cite

Cite as: Jiajun Wu, Jian Yang, Wei Zhang, Linzheng Chai, Yuchi Ma, Ensheng Shi, Yuqing Ma, Zhoujun Li and Xianglong Liu. 2026. UCoder: Unsupervised Code Generation by Internal Probing of Large Language Models. Findings of the Association for Computational Linguistics: ACL 2026.

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

```bibtex
@inproceedings{wu2026ucoder,
  title = {{UCoder: Unsupervised Code Generation by Internal Probing of Large Language Models}},
  author = {Jiajun Wu, Jian Yang, Wei Zhang, Linzheng Chai, Yuchi Ma, Ensheng Shi, Yuqing Ma, Zhoujun Li and Xianglong Liu},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.277},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.277}
}
```

- [← previous T²-RAGBench: Text-and-Table Benchmark for Evaluating Retrieval-Augmented Generation](https://md.research.elamur.ai/free/en/papers/strich-2026-t-2-ragbench-text-table/)
- [next → UI-Copilot: Advancing Long-Horizon GUI Automation via Tool-Integrated Policy Optimization](https://md.research.elamur.ai/free/en/papers/lu-2026-ui-copilot-advancing-long-horizon/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/18/)

## References

34 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Anthropic. 2025. Introducing claude sonnet 4.5. [find](https://scholar.google.com/scholar?q=Anthropic%2E+2025%2E+Introducing+claude+sonnet+4%2E5%2E)
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. Federico Cassano, John Gouwar, Daniel Nguyen, Sydney Nguyen, Luna Phipps-Costin, Donald Pinckney, Ming-Ho Yee, Yangtian Zi, Carolyn Jane Anderson, Molly Q Feldman, Arjun Guha, Michael Greenberg, and Abhinav Jangda. 2023. Multipl-e: A scalable and polyglot approach to benchmarking neural code generat… [find](https://scholar.google.com/scholar?q=Federico+Cassano%2C+John+Gouwar%2C+Daniel+Nguyen%2C+Sydney+Nguyen%2C+Luna+Phipps%2DCostin%2C+Donald+Pinckney%2C+Ming%2DHo+Yee%2C+Yangtian+Zi%2C+Carolyn+Jane+Anderson%2C+Molly+Q+Feldman%2C+Arjun+Guha%2C+Michael+Greenberg%2C+and+A)
4. Linzheng Chai, Shukai Liu, Jian Yang, Yuwei Yin, Ke Jin, Jiaheng Liu, Tao Sun, Ge Zhang, Changyu Ren, Hongcheng Guo, et al. 2024. Mceval: Massively multilingual code evaluation. arXiv preprint arXiv:2406.07436. — [arXiv:2406.07436](https://arxiv.org/abs/2406.07436) [find](https://scholar.google.com/scholar?q=Linzheng+Chai%2C+Shukai+Liu%2C+Jian+Yang%2C+Yuwei+Yin%2C+Ke+Jin%2C+Jiaheng+Liu%2C+Tao+Sun%2C+Ge+Zhang%2C+Changyu+Ren%2C+Hongcheng+Guo%2C+et+al%2E+2024%2E+Mceval%3A+Massively+multilingual+code+evaluation%2E+arXiv+preprint+arXiv%3A2)
5. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde 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, Nick Ryder, Mikhai… — [Evaluating Large Language Models Trained on Code](https://md.research.elamur.ai/free/en/papers/chen-2021-evaluating-large-language-models-trained/)
6. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Pondé de Oliveira Pinto, Jared Kaplan, Harrison 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, Nick Ryder, Mik… — [Evaluating Large Language Models Trained on Code](https://md.research.elamur.ai/free/en/papers/chen-2021-evaluating-large-language-models-trained/)
7. Zhuo, Thomas Wang, Olivier Dehaene, Mishig Davaadorj, Joel Lamy-Poirier, João Monteiro, Oleh Shliazhko, Nicolas Gontier, Nicholas Meade, Armel Zebaze, Ming-Ho Yee, Logesh Kumar Umapathi, Jian Zhu, Benjamin Lipkin, Muhtasham Oblokulov, Zhiruo Wang, Rudra Murthy V, Jason Stillerman, Siva Sankalp Patel… [find](https://scholar.google.com/scholar?q=Zhuo%2C+Thomas+Wang%2C+Olivier+Dehaene%2C+Mishig+Davaadorj%2C+Joel+Lamy%2DPoirier%2C+Jo%C3%A3o+Monteiro%2C+Oleh+Shliazhko%2C+Nicolas+Gontier%2C+Nicholas+Meade%2C+Armel+Zebaze%2C+Ming%2DHo+Yee%2C+Logesh+Kumar+Umapathi%2C+Jian+Zhu%2C+Ben)
8. Tianzhe Chu, Yuexiang Zhai, Jihan Yang, Shengbang Tong, Saining Xie, Dale Schuurmans, Quoc V Le, Sergey Levine, and Yi Ma. 2025. [Sft memorizes, rl generalizes: A comparative study of foundation model post-training](https://md.research.elamur.ai/free/en/papers/chu-2025-sft-memorizes-rl-generalizes-comparative/). arXiv preprint arXiv:2501.17161.
9. Daya Guo, Qihao Zhu, Dejian Yang, Zhenda Xie, Kai Dong, Wentao Zhang, Guanting Chen, Xiao Bi, Y Wu, YK Li, et al. 2024. Deepseek-coder: When the large language model meets programming–the rise of code intelligence. arXiv preprint arXiv:2401.14196. [find](https://scholar.google.com/scholar?q=Daya+Guo%2C+Qihao+Zhu%2C+Dejian+Yang%2C+Zhenda+Xie%2C+Kai+Dong%2C+Wentao+Zhang%2C+Guanting+Chen%2C+Xiao+Bi%2C+Y+Wu%2C+YK+Li%2C+et+al%2E+2024%2E+Deepseek%2Dcoder%3A+When+the+large+language+model+meets+programming%E2%80%93the+rise+of+code)
10. Siming Huang, Tianhao Cheng, Jason Klein Liu, Weidi Xu, Jiaran Hao, Liuyihan Song, Yang Xu, Jian Yang, Jiaheng Liu, Chenchen Zhang, Linzheng Chai, Ruifeng Yuan, Xianzhen Luo, Qiufeng Wang, Yuan-Tao Fan, Qingfu Zhu, Zhaoxiang Zhang, Yang Gao, Jie Fu, Qian Liu, Houyi Li, Ge Zhang, Yuan Qi, Yinghui Xu,… — [OpenCoder: The Open Cookbook for Top-Tier Code Large Language Models](https://md.research.elamur.ai/free/en/papers/huang-2025-opencoder-open-cookbook-top-tier/)
11. Binyuan Hui, Jian Yang, Zeyu Cui, Jiaxi Yang, Dayiheng Liu, Lei Zhang, Tianyu Liu, Jiajun Zhang, Bowen Yu, Kai Dang, et al. 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/)
12. 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/). CoRR, abs/2403.07974.
13. Peng Lai, Zhihao Ou, Yong Wang, Longyue Wang, Jian Yang, Yun Chen, and Guanhua Chen. 2026. Biass-cope: Towards automated detection of bias in LLM-as-a-judge evaluation. In The Fourteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Peng+Lai%2C+Zhihao+Ou%2C+Yong+Wang%2C+Longyue+Wang%2C+Jian+Yang%2C+Yun+Chen%2C+and+Guanhua+Chen%2E+2026%2E+Biass%2Dcope%3A+Towards+automated+detection+of+bias+in+LLM%2Das%2Da%2Djudge+evaluation%2E+In+The+Fourteenth+International)
14. Lehui Li, Ruining Wang, Haochen Song, Yaoxin Mao, Tong Zhang, Yuyao Wang, Jiayi Fan, Yitong Zhang, Jieping Ye, Chengqi Zhang, and Yongshun Gong. 2026. What papers don’t tell you: Recovering tacit knowledge for automated paper reproduction. [find](https://scholar.google.com/scholar?q=Lehui+Li%2C+Ruining+Wang%2C+Haochen+Song%2C+Yaoxin+Mao%2C+Tong+Zhang%2C+Yuyao+Wang%2C+Jiayi+Fan%2C+Yitong+Zhang%2C+Jieping+Ye%2C+Chengqi+Zhang%2C+and+Yongshun+Gong%2E+2026%2E+What+papers+don%E2%80%99t+tell+you%3A+Recovering+tacit+know)
15. Raymond Li, Loubna Ben Allal, Yangtian Zi, Niklas Muennighoff, Denis Kocetkov, Chenghao Mou, Marc Marone, Christopher Akiki, Jia Li, Jenny Chim, Qian Liu, Evgenii Zheltonozhskii, Terry Yue Yixia Li, Boya Xiong, Guanhua Chen, and Yun Chen. 2024. Setar: Out-of-distribution detection with selective low… [find](https://scholar.google.com/scholar?q=Raymond+Li%2C+Loubna+Ben+Allal%2C+Yangtian+Zi%2C+Niklas+Muennighoff%2C+Denis+Kocetkov%2C+Chenghao+Mou%2C+Marc+Marone%2C+Christopher+Akiki%2C+Jia+Li%2C+Jenny+Chim%2C+Qian+Liu%2C+Evgenii+Zheltonozhskii%2C+Terry+Yue+Yixia+Li%2C+B)
16. Shukai Liu, Jian Yang, Bo Jiang, Yizhi Li, Jinyang Guo, Xianglong Liu, and Bryan Dai. 2025. Context as a tool: Context management for long-horizon swe-agents. [find](https://scholar.google.com/scholar?q=Shukai+Liu%2C+Jian+Yang%2C+Bo+Jiang%2C+Yizhi+Li%2C+Jinyang+Guo%2C+Xianglong+Liu%2C+and+Bryan+Dai%2E+2025%2E+Context+as+a+tool%3A+Context+management+for+long%2Dhorizon+swe%2Dagents%2E)
17. Siyao Liu, He Zhu, Jerry Liu, Shulin Xin, Aoyan Li, Rui Long, Li Chen, Jack Yang, Jinxiang Xia, ZY Peng, et al. 2024. Fullstack bench: Evaluating llms as full stack coder. arXiv preprint arXiv:2412.00535. — [arXiv:2412.00535](https://arxiv.org/abs/2412.00535) [find](https://scholar.google.com/scholar?q=Siyao+Liu%2C+He+Zhu%2C+Jerry+Liu%2C+Shulin+Xin%2C+Aoyan+Li%2C+Rui+Long%2C+Li+Chen%2C+Jack+Yang%2C+Jinxiang+Xia%2C+ZY+Peng%2C+et+al%2E+2024%2E+Fullstack+bench%3A+Evaluating+llms+as+full+stack+coder%2E+arXiv+preprint+arXiv%3A2412%2E00)
18. Anton Lozhkov, Raymond Li, Loubna Ben Allal, Federico Cassano, Joel Lamy-Poirier, Nouamane Tazi, Ao Tang, Dmytro Pykhtar, Jiawei Liu, Yuxiang Wei, et al. 2024. Starcoder 2 and the stack v2: The next generation. arXiv preprint arXiv:2402.19173. [find](https://scholar.google.com/scholar?q=Anton+Lozhkov%2C+Raymond+Li%2C+Loubna+Ben+Allal%2C+Federico+Cassano%2C+Joel+Lamy%2DPoirier%2C+Nouamane+Tazi%2C+Ao+Tang%2C+Dmytro+Pykhtar%2C+Jiawei+Liu%2C+Yuxiang+Wei%2C+et+al%2E+2024%2E+Starcoder+2+and+the+stack+v2%3A+The+next+g)
19. Ziyang Luo, Can Xu, Pu Zhao, Qingfeng Sun, Xiubo Geng, Wenxiang Hu, Chongyang Tao, Jing Ma, Qingwei Lin, and Daxin Jiang. 2023. Wizardcoder: Empowering code large language models with evol-instruct. CoRR, abs/2306.08568. [find](https://scholar.google.com/scholar?q=Ziyang+Luo%2C+Can+Xu%2C+Pu+Zhao%2C+Qingfeng+Sun%2C+Xiubo+Geng%2C+Wenxiang+Hu%2C+Chongyang+Tao%2C+Jing+Ma%2C+Qingwei+Lin%2C+and+Daxin+Jiang%2E+2023%2E+Wizardcoder%3A+Empowering+code+large+language+models+with+evol%2Dinstruct%2E+C)
20. OpenAI. 2025. Introducing upgrades to codex: Gpt-5-codex. [https://openai.com/index/](https://openai.com/index/) introducing-upgrades-to-codex/. [↗](https://openai.com/index/)
21. Mikhail Pravilov, Egor Bogomolov, Yaroslav Golubev, and Timofey Bryksin. 2021. Unsupervised learning of general-purpose embeddings for code changes. In Proceedings of the 5th international workshop on machine learning techniques for software quality evolution, pages 7–12. [find](https://scholar.google.com/scholar?q=Mikhail+Pravilov%2C+Egor+Bogomolov%2C+Yaroslav+Golubev%2C+and+Timofey+Bryksin%2E+2021%2E+Unsupervised+learning+of+general%2Dpurpose+embeddings+for+code+changes%2E+In+Proceedings+of+the+5th+international+workshop+on)
22. Baptiste Rozière, Jonas Gehring, Fabian Gloeckle, Sten Sootla, Itai Gat, Xiaoqing Ellen Tan, Yossi Adi, Jingyu Liu, Tal Remez, Jérémy Rapin, Artyom Kozhevnikov, Ivan Evtimov, Joanna Bitton, Manish Bhatt, Cristian Canton-Ferrer, Aaron Grattafiori, Wenhan Xiong, Alexandre Défossez, Jade Copet, Faisal… — [Code Llama: Open Foundation Models for Code](https://md.research.elamur.ai/free/en/papers/roziere-2024-code-llama-open-foundation-models/)
23. Yuxiang Wei, Zhe Wang, Jiawei Liu, Yifeng Ding, and Lingming Zhang. 2023. Magicoder: Source code is all you need. CoRR, abs/2312.02120. [find](https://scholar.google.com/scholar?q=Yuxiang+Wei%2C+Zhe+Wang%2C+Jiawei+Liu%2C+Yifeng+Ding%2C+and+Lingming+Zhang%2E+2023%2E+Magicoder%3A+Source+code+is+all+you+need%2E+CoRR%2C+abs%2F2312%2E02120%2E)
24. Jian Yang, Xianglong Liu, Weifeng Lv, Ken Deng, Shawn Guo, Lin Jing, Yizhi Li, Shark Liu, Xianzhen Luo, Yuyu Luo, Changzai Pan, Ensheng Shi, Yingshui Tan, Renshuai Tao, Jiajun Wu, Xianjie Wu, Zhenhe Wu, Daoguang Zan, Chenchen Zhang, Wei Zhang, He Zhu, Terry Yue Zhuo, Kerui Cao, Xianfu Cheng, Jun Don… [find](https://scholar.google.com/scholar?q=Jian+Yang%2C+Xianglong+Liu%2C+Weifeng+Lv%2C+Ken+Deng%2C+Shawn+Guo%2C+Lin+Jing%2C+Yizhi+Li%2C+Shark+Liu%2C+Xianzhen+Luo%2C+Yuyu+Luo%2C+Changzai+Pan%2C+Ensheng+Shi%2C+Yingshui+Tan%2C+Renshuai+Tao%2C+Jiajun+Wu%2C+Xianjie+Wu%2C+Zhenhe+W)
25. Yin. 2023. Wavecoder: Widespread and versatile enhanced instruction tuning with refined data generation. CoRR, abs/2312.14187. — [arXiv:2312.14187](https://arxiv.org/abs/2312.14187) [find](https://scholar.google.com/scholar?q=Yin%2E+2023%2E+Wavecoder%3A+Widespread+and+versatile+enhanced+instruction+tuning+with+refined+data+generation%2E+CoRR%2C+abs%2F2312%2E14187%2E+%E2%80%94+%5BarXiv%3A2312%2E14187%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2312%2E14187%29)
26. Yang Yue, Zhiqi Chen, Rui Lu, Andrew Zhao, Zhaokai Wang, Yang Yue, Shiji Song, and Gao Huang. [Does reinforcement learning really incentivize reasoning capacity in llms beyond the base model?](https://md.research.elamur.ai/free/en/papers/yue-2025-reinforcement-learning-really-incentivize-reasoning/), 2025. URL https://arxiv.org/abs/2504.13837. [↗](https://arxiv.org/abs/2504.13837)
27. Wei Zhang, Jack Yang, Renshuai Tao, Lingzheng Chai, Shawn Guo, Jiajun Wu, Xiaoming Chen, Ganqu Cui, Ning Ding, Xander Xu, Hu Wei, and Bowen Zhou. 2025. [V-gamegym: Visual game generation for code large language models](https://md.research.elamur.ai/free/en/papers/zhang-2026-v-gamegym-visual-game-generation/).
28. Yitong Zhang, Chengze Li, Ruize Chen, Guowei Yang, Xiaoran Jia, Yijie Ren, and Jia Li. 2026. To see is not to master: Teaching llms to use private libraries for code generation. [find](https://scholar.google.com/scholar?q=Yitong+Zhang%2C+Chengze+Li%2C+Ruize+Chen%2C+Guowei+Yang%2C+Xiaoran+Jia%2C+Yijie+Ren%2C+and+Jia+Li%2E+2026%2E+To+see+is+not+to+master%3A+Teaching+llms+to+use+private+libraries+for+code+generation%2E)
29. Terry Yue Zhuo, Minh Chien Vu, Jenny Chim, Han Hu, Wenhao Yu, Ratnadira Widyasari, Imam Nur Bani Yusuf, Haolan Zhan, Junda He, Indraneil Paul, et al. 2024. [Bigcodebench: Benchmarking code generation with diverse function calls and complex instructions](https://md.research.elamur.ai/free/en/papers/zhuo-2024-bigcodebench-benchmarking-code-generation-diverse/). arXiv preprint arXiv:2406.15877.
30. Jian Yang, Jiaxi Yang, Ke Jin, Yibo Miao, Lei Zhang, Liqun Yang, Zeyu Cui, Yichang Zhang, Binyuan Hui, and Junyang Lin. 2024. Evaluating and aligning codellms on human preference. arXiv preprint arXiv:2412.05210. — [arXiv:2412.05210](https://arxiv.org/abs/2412.05210) [find](https://scholar.google.com/scholar?q=Jian+Yang%2C+Jiaxi+Yang%2C+Ke+Jin%2C+Yibo+Miao%2C+Lei+Zhang%2C+Liqun+Yang%2C+Zeyu+Cui%2C+Yichang+Zhang%2C+Binyuan+Hui%2C+and+Junyang+Lin%2E+2024%2E+Evaluating+and+aligning+codellms+on+human+preference%2E+arXiv+preprint+arXiv)
31. Jian Yang, Wei Zhang, Shawn Guo, Zhengmao Ye, Lin Jing, Shark Liu, Yizhi Li, Jiajun Wu, Cening Liu, X Ma, et al. 2026a. Iquest-coder-v1 technical report. arXiv preprint arXiv:2603.16733. — [arXiv:2603.16733](https://arxiv.org/abs/2603.16733) [find](https://scholar.google.com/scholar?q=Jian+Yang%2C+Wei+Zhang%2C+Shawn+Guo%2C+Zhengmao+Ye%2C+Lin+Jing%2C+Shark+Liu%2C+Yizhi+Li%2C+Jiajun+Wu%2C+Cening+Liu%2C+X+Ma%2C+et+al%2E+2026a%2E+Iquest%2Dcoder%2Dv1+technical+report%2E+arXiv+preprint+arXiv%3A2603%2E16733%2E+%E2%80%94+%5BarXiv%3A2603)
32. Jian Yang, Wei Zhang, Jiajun Wu, Junhang Cheng, Shawn Guo, Haowen Wang, Weicheng Gu, Yaxin Du, Joseph Li, Fanglin Xu, et al. 2026b. Incoder-32b: Code foundation model for industrial scenarios. arXiv preprint arXiv:2603.16790. [find](https://scholar.google.com/scholar?q=Jian+Yang%2C+Wei+Zhang%2C+Jiajun+Wu%2C+Junhang+Cheng%2C+Shawn+Guo%2C+Haowen+Wang%2C+Weicheng+Gu%2C+Yaxin+Du%2C+Joseph+Li%2C+Fanglin+Xu%2C+et+al%2E+2026b%2E+Incoder%2D32b%3A+Code+foundation+model+for+industrial+scenarios%2E+arXiv+p)
33. Jian Yang, Wei Zhang, Jiajun Wu, Junhang Cheng, Tuney Zheng, Fanglin Xu, Weicheng Gu, Lin Jing, Yaxin Du, Joseph Li, et al. 2026c. Incoder-32b-thinking: Industrial code world model for thinking. arXiv preprint arXiv:2604.03144. [find](https://scholar.google.com/scholar?q=Jian+Yang%2C+Wei+Zhang%2C+Jiajun+Wu%2C+Junhang+Cheng%2C+Tuney+Zheng%2C+Fanglin+Xu%2C+Weicheng+Gu%2C+Lin+Jing%2C+Yaxin+Du%2C+Joseph+Li%2C+et+al%2E+2026c%2E+Incoder%2D32b%2Dthinking%3A+Industrial+code+world+model+for+thinking%2E+arXiv)
34. Junjie Ye, Yuming Yang, Yang Nan, Shuo Li, Qi Zhang, Tao Gui, Xuan-Jing Huang, Peng Wang, Zhongchao Shi, and Jianping Fan. 2025. Analyzing the effects of supervised fine-tuning on model knowledge from token and parameter levels. In Proceedings of the 2025 Conference on Empirical Methods in Natural L… [find](https://scholar.google.com/scholar?q=Junjie+Ye%2C+Yuming+Yang%2C+Yang+Nan%2C+Shuo+Li%2C+Qi+Zhang%2C+Tao+Gui%2C+Xuan%2DJing+Huang%2C+Peng+Wang%2C+Zhongchao+Shi%2C+and+Jianping+Fan%2E+2025%2E+Analyzing+the+effects+of+supervised+fine%2Dtuning+on+model+knowledge+from)

*Updated 31 October 2026*