Jialiang Gu

George Mason University

Fairfax, VA, USA

Hi, I’m Jialiang Gu (谷嘉良), a second-year Ph.D. student in Computer Science at George Mason University, advised by Prof. Keren Zhou. In Summer 2026, I was a Research Intern at Microsoft Research in Redmond, where I worked with Debadeepta Dey and Mason Remy. I also collaborate closely with Prof. Jialu Zhang at the University of Waterloo.

My research focuses on AI systems and compilers, particularly efficient LLM inference, GPU programming systems, and program analysis for reliable AI software. I am interested in building tools that bridge high-level AI models and low-level execution, including compiler/runtime optimization, performance analysis, and formal reasoning about AI programs.

I spent quite a few years solving algorithmic puzzles competitively. Along the way, I won a Gold Medal at the ACM-ICPC Asia Regional Contest and became the 2024 Hubei Provincial ICPC Champion.

I’m always open to conversations and collaborations around these topics. Feel free to reach out.

news

Jul 05, 2026	I will attend FSE 2026 to present our FSE-IVR paper. See you in Montréal!
May 18, 2026	I will join Microsoft Research AIFX Group as research intern! See you in Redmond.
Apr 01, 2026	One first author paper, “Context-Aware Feedback Compression in Online Judge Programming with LLMs,” was accepted to FSE-IVR 2026!
Aug 27, 2025	I have joined GMU as a CS PhD student, advised by Keren Zhou!

selected publications

AACL-IJCNLP

Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline

Zhikai Chen^*, Jialiang Gu^*, Junyu Yin^*, Xianxuan Long, Shenglai Zeng, Xiaoze Liu, Kai Guo, Keren Zhou, and Jiliang Tang

In Asia-Pacific Chapter of the Association for Computational Linguistics and International Joint Conference on Natural Language Processing, Nov 2026

Accepted to AACL-IJCNLP 2026; * Equal contribution (co-first authors)

arXiv Bib HTML PDF

@inproceedings{chen2026automem,
  title = {Exploring Cross-Scenario Generality of Agentic Memory Systems: Diagnostics and a Strong Baseline},
  author = {Chen, Zhikai and Gu, Jialiang and Yin, Junyu and Long, Xianxuan and Zeng, Shenglai and Liu, Xiaoze and Guo, Kai and Zhou, Keren and Tang, Jiliang},
  booktitle = {Asia-Pacific Chapter of the Association for Computational Linguistics and International Joint Conference on Natural Language Processing},
  year = {2026},
  month = nov,
  category = {publication},
  note = {Accepted to AACL-IJCNLP 2026; * Equal contribution (co-first authors)}
}

FSE-IVR

Context-Aware Feedback Compression in Online Judge Programming with LLMs

Jialiang Gu, Keren Zhou, Daming Li, Hanyuan Shi, and Jialu Zhang

In Proceedings of the 34th ACM International Conference on the Foundations of Software Engineering, Ideas, Visions and Reflections Track, Apr 2026

Accepted to FSE-IVR 2026

Bib

@inproceedings{gu2026context,
  title = {Context-Aware Feedback Compression in Online Judge Programming with LLMs},
  author = {Gu, Jialiang and Zhou, Keren and Li, Daming and Shi, Hanyuan and Zhang, Jialu},
  booktitle = {Proceedings of the 34th ACM International Conference on the Foundations of Software Engineering, Ideas, Visions and Reflections Track},
  year = {2026},
  month = apr,
  category = {publication},
  note = {Accepted to FSE-IVR 2026}
}

SIGCSE TS

NKI Teaching Lab: An Accelerator-Free Toolchain for Learning Tile-Level AI Accelerator Programming

Jialiang Gu, Junjie Ma, Thai-Hoa Nguyen, Hao Wu, Zhicong Lu, Shuyin Jiao, Binqian Yin, and Keren Zhou

In ACM Technical Symposium on Computer Science Education, Feb 2027

Accepted to SIGCSE TS 2027

Bib HTML

@inproceedings{gu2027nkilab,
  title = {{NKI} Teaching Lab: An Accelerator-Free Toolchain for Learning Tile-Level {AI} Accelerator Programming},
  author = {Gu, Jialiang and Ma, Junjie and Nguyen, Thai-Hoa and Wu, Hao and Lu, Zhicong and Jiao, Shuyin and Yin, Binqian and Zhou, Keren},
  booktitle = {ACM Technical Symposium on Computer Science Education},
  year = {2027},
  month = feb,
  category = {publication},
  note = {Accepted to SIGCSE TS 2027}
}