News
-
January 23, 2025
One paper will be presented at NAACL 2025. Congrats to Xiangyan Liu, and our collaborators!
CodexGraph: Bridging Large Language Models and Code Repositories via Code Graph Databases -
January 23, 2025
Three papers will be presented at ICLR 2025. Congrats to Jinjie Ni, Yiran Zhao, Guanzheng Chen, and our collaborators!
MixEval-X: Any-to-Any Evaluations from Real-World Data Mixtures
Identifying and Tuning Safety Neurons in Large Language Models
LongPO: Long Context Self-Evolution of Large Language Models through Short-to-Long Preference Optimization -
December 19, 2024
One paper will be presented at AAAI 2025. Congrats to Leon Lin, Hannah Brown and our collaborators!
Single Character Perturbations Break LLM Alignment
Featured Research
Safety Neuron
We propose a neuron detection method to identify and tune safety neurons—less than 1% of model parameters—using SN-Tune, which enhances safety without affecting general model performance.
Noisy Student
The first method that uses extra unlabeled data to achieve state-of-the-art on ImageNet. Also employed in AlphaFold 2 (Nobel Prize 2024), Google Search and other state-of-the-art AI systems.
Mission
Feel the AGI. Use 10k GPUs like rich kids in the industry. Get 100k citations. Get Nobel prizes and Turing awards.
Just kidding haha. We work on democratizing AI in aspects including but not limited to resource-efficient AI, vision language models and more exploratory topics.
People
Faculty
Michael Qizhe Shieh
Assistant Professor
Jinjie Ni
Research Fellow
PhD Students
Hannah Brown
Co-advised with Kenji Kawaguchi
Esther Gan
Co-advised with Min-Yen Kan
Yiran Zhao
Affiliated Member
Masters Students
Fengtao He
National University of Singapore
Leon Lin
National University of Singapore
Undergraduate Students
Qilong Feng
National University of Singapore
Zhennan Shen
Shanghai Jiao Tong University
Yang Zhang
Peking University
Publications
*: equal contribution, †: equal advising
