Hello! I’m a master’s student in Computer Science and Technology at Harbin Institute of Technology, where I work with the Machine Intelligence and Translation Lab (Language Technology Research Center), advised by Prof. Tiejun Zhao.

My research interests include Multimodal and Multilingual LLMs, Multi-Agent Systems, and Multimodal Reasoning. I’m particularly interested in how language models can reason, collaborate, and interact with the world.

Beyond research, I’m an open-source enthusiast and enjoy exploring Linux and the broader software ecosystem.

🔥 News

2026.09 — 🎉 Released RepoMAS, an issue-driven framework for progressively specified tasks. Submitted to ICLR 2027. Paper
2026.02 — 🎉 Our paper C³B was accepted to ICLR 2026! Paper
2025.02 — 📢 Released a comprehensive evaluation of o1-like LLMs for multilingual translation. Paper

📝 Publications

RepoMAS: Solving Progressively Specified Tasks with Issue-Driven Multi-Agent Systems

Yuchen Song^*, Andong Chen^*, Wenxin Zhu, Muyun Yang, Tiejun Zhao

Under review at ICLR 2027 · Paper
Culture In a Frame: C³B as a Comic-Based Benchmark for Multimodal Culturally Awareness

Yuchen Song^*, Andong Chen^*, Wenxin Zhu, Kehai Chen, Xuefeng Bai, Muyun Yang, Tiejun Zhao

ICLR 2026 · Paper
Make Imagination Clearer! Stable Diffusion-based Visual Imagination for Multimodal Machine Translation

Andong Chen, Yuchen Song, Kehai Chen, Xuefeng Bai, Muyun Yang, Liqiang Nie, Jie Liu, Tiejun Zhao, Min Zhang

ACL 2025 · Paper
Evaluating o1-Like LLMs: Unlocking Reasoning for Translation through Comprehensive Analysis

Andong Chen^*, Yuchen Song^*, Wenxin Zhu^*, Kehai Chen, Muyun Yang, Tiejun Zhao, Min Zhang

arXiv preprint · Paper

Cited in Inverse Scaling in Test-Time Compute (TMLR 2025, Anthropic Fellows Program).
Thinking with Comics: Enhancing Multimodal Reasoning through Structured Visual Storytelling

Andong Chen, Wenxin Zhu, Qiuyu Ding, Yuchen Song, Muyun Yang, Tiejun Zhao

Under review at AAAI 2027 · Paper

^* Equal contribution.

🎖 Honors & Awards

Special Prize Academic Scholarship (multiple times), Faculty of Computing, Harbin Institute of Technology
Shengshengbuxi Scholarship (2nd edition), Harbin Institute of Technology
University Scholarship (Top 10%, multiple times), Harbin Institute of Technology

📖 Education

2025.09 – 2028.03 · M.S. in Computer Science and Technology, Harbin Institute of Technology
- Machine Intelligence and Translation Lab
2021.09 – 2025.06 · B.Eng. in Artificial Intelligence, Harbin Institute of Technology
- School of Future Technology
- Co-founder of the HIT Linux Open Source Student Club

🛠 What I Work With

Most of my work revolves around LLMs and multi-agent systems. I primarily use Python, along with PyTorch and the Hugging Face ecosystem, for model training, fine-tuning, and evaluation. I also work with C/C++ and enjoy exploring algorithms and how things work under the hood.

I’m particularly interested in LLM post-training, from supervised fine-tuning and parameter-efficient methods like LoRA to reinforcement learning with reward models and verifiable feedback.

On the agent side, I enjoy exploring how LLMs can plan, use tools, collaborate, and verify their own work. I’ve been working with multi-agent architectures, task decomposition, and tool integration, including MCP and agent skills.

I usually develop in Linux environments, manage experiments and code with Git/GitHub, and care about reproducibility, thoughtful evaluation, and clean implementations.

🌱 Misc.