Qunhong Zeng

Beijing, China

I am a final-year graduate student at the School of Computer Science and Technology at Beijing Institute of Technology, advised by Prof. Yuxia Zhang. Prior to this, I received my B.S. degree in Computer Science from the same university in 2023.

My research primarily focuses on AI for Software Engineering (AI4SE). I am currently working at MiniMax AI, where I focus on improving foundation models’ agentic coding capabilities and scaling verifiable environments for agentic post-training. My recent work centers on SWE tasks (Multi-SWE-bench, SWE-bench multilingual, etc.).

I am also interested in MLSys and enjoy building scalable systems, with experience in building RL infrastructure (e.g., verl) and scaling agent scaffolds.

Please feel free to contact me if you’re interested in having a discussion :)

Publications

OctoBench: Benchmarking Scaffold-Aware Instruction Following in Repository-Grounded Agentic Coding

Deming Ding, Shichun Liu, Enhui Yang, Jiahang Lin, Ziying Chen, Shihan Dou, Honglin Guo, Weiyu Cheng, Pengyu Zhao, Chengjun Xiao, Qunhong Zeng, Qi Zhang, Xuanjing Huang, Qidi Xu, and Tao Gui

arXiv preprint arXiv:2601.10343, 2026

PDF Code
ToolTrain: Tool-integrated Reinforcement Learning for Repo Deep Search

Zexiong Ma, Chao Peng, Qunhong Zeng, Pengfei Gao, Yanzhen Zou, and Bing Xie

In Proceedings of the IEEE/ACM 48th International Conference on Software Engineering (ICSE), 2026

PDF Code
Evaluating Generated Commit Messages with Large Language Models

Qunhong Zeng, Yuxia Zhang, Zexiong Ma, Bo Jiang, Ningyuan Sun, Klaas-Jan Stol, Xingyu Mou, and Hui Liu

In Proceedings of the IEEE/ACM 48th International Conference on Software Engineering (ICSE), 2026

PDF Code
A First Look at Conventional Commits Classification

Qunhong Zeng, Yuxia Zhang, Zhiqing Qiu, and Hui Liu

In Proceedings of the IEEE/ACM 47th International Conference on Software Engineering (ICSE), 2025

PDF Code
COLARE: Commit Classification via Fine-grained Context-aware Representation of Code Changes

Qunhong Zeng, Yuxia Zhang, Zeyu Sun, Yujie Guo, and Hui Liu

In 2024 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER), 2024

PDF Code

Experiences

MiniMax, Top Talent Intern, 2024.7 - now
- Working on MiniMax-M2.x and more, stay tuned ~
- Core contributor to improving the SWE/MSWE coding and review capabilities of MiniMax-M2.1. Created the swe-review benchmark, to be open-sourced soon.
- Core contributor to MiniMax-M2’s multilingual coding capabilities. Benefiting from environment scaling and scaffold scaling, MiniMax-M2 shows SOTA performance on Multi-SWE-bench and SWE-bench multilingual among all open-source models.
Meituan, Beidou Program Intern, 2025.6 - 2025.7
- Researched multi-turn conversational RL to enhance LLM’s dialogue capabilities.
ByteDance, Trae CodeAI Intern, 2024.9 - 2025.6
- Conducted research on automated commit message generation and quality evaluation, with findings accepted for publication in the research track of ICSE 2026.
- Researched RL approaches to enhance LLM’s reasoning capabilities, such as mathematical reasoning, competitive programming contest test generation, and tool-interactive reasoning. Collaborated with Zexiong Ma and mentored by Bo Jiang.
Oneflow, Framework Intern, 2024.3 - 2024.9
- Extended deep learning framework oneflow compatibility beyond CUDA to diverse hardware accelerators (NPU/XPU), implementing Ascend CANN/AscendC-based operators that enabled end-to-end ResNet, GPT2, and Llama model inference/training on Ascend chips.
- Managed framework infrastructure including Docker containerization, CI/CD, cross-platform compilation, and Manylinux-compliant Python/C++ extension packaging.
Momenta, HD-Map Backend Intern, 2022.9 - 2023.1
- Engineered high-performance Python/C++ extensions for HD maps algorithm library, delivering 20x performance improvement.
ByteDance, AI-Lab Intern, 2021.12 - 2022.6
- Contributed to ByteDance’s machine learning platform development as part of the engineering team.