About me

Hi, there!

I’m Yuhong Luo, a third-year PhD student in Computer Science at Rutgers.

Currently, I am working with Xintong Wang on the intersaction of AI and algorithmic game theory.

On a higher level, my research interest is in AI robustness, safety and trustworthiness. Particularly, I am interested in studying AI systems (e.g., multi-agent LLMs) from both the algorithmic and statistical perspectives.

Before Rutgers, I was a Master’s student at UMass, working with Philip S. Thomas and Przemek Grabowicz on providing machine learning algorithms with high-confidence safety and fairness guarantees and with Pan Li on scalable and neighborhood-aware temporal graph representation learning. I was also a software engineer working on marketing technology and promotional content management at Airbnb.

I am happy to connect and chat! Feel free to reach me at y.luo@rutgers.edu!

I am actively looking for AI/ML research internship opportunities for summer 2027. Please feel free to reach out with relevant opportunities!

🔥 News in 2026

First-author paper accepted by NeurIPS 2026!
- Decentralized Aggregation of LLM Predictions via Wagering Mechanisms
First-author paper accepted by AAMAS 2026!
- Algorithmic Collusion at Test Time: A Meta-game Design and Evaluation

Past news

- First-author paper accepted by **NeurIPS** 2025! - [Fair Representation Learning with Controllable High Confidence Guarantees via Adversarial Inference](https://arxiv.org/abs/2510.21017) - First-author paper accepted by **Learning on Graphs** 2024! - [Scalable and Efficient Temporal Graph Representation Learning via Forward Recent Sampling](https://arxiv.org/abs/2402.01964) - First-author paper accepted by **Learning on Graphs** 2022 and received the **best paper award** (1 of 2)! - [Neighborhood-aware Scalable Temporal Network Representation Learning](https://arxiv.org/abs/2209.01084)