Director
Rex Ying
We are a group of data-driven machine learning enthusiasts who are primarily interested in building unified approaches to integrate and learn from complex real-world data. Beyond just text and images, we also build novel deep learning models that consider graphs, time series, geometry and tabular data, and use them to solve a wide array of applications in domains such as biology, medicine, chemistry, physics, neuroscience, social networks, science of science and supply chain. Motivated by real-world use cases, we focus on efficient and scalable techniques that combine relational reasoning, multimodal learning, geometric deep learning and foundation models. Furthermore, we are actively doing research in trustworthy deep learning to allow safe, transparent and reliable deployment of such models.
🏠 Homepage
Postdocs
Haiwen Wang
My research focuses on multimodal time series learning and foundation model development, particularly on integrating sequential, textual, and structured signals for real-world decision-making. I am particularly interested in Transformers, large language models, and reinforcement learning, as well as in how foundation models can better understand, forecast, and act on complex temporal data.
🏠 Homepage
PhD Students
Tinglin Huang
My research interests revolve around computational biology, including macromolecule modeling and geometric deep learning on 3D molecular structure.
🏠 Homepage
Jialin Chen
My research focuses on advancing multimodal learning techniques to seamlessly integrate heterogeneous data modalities, with a focus on leveraging linguistic information to enhance graph representation learning. I also explore cutting-edge approaches to develop robust graph foundation models and scalable pretraining strategies, pushing the boundaries of AI’s ability to reason over complex and structured data. Alongside these directions, I’m also committed to advancing Trustworthy AI that aims to address the explainability and reliability concerns associated with large models (e.g., LLMs) and real-world applications.
🏠 Homepage
Weikang Qiu
My research focuses on the intersection of machine learning and neuroscience. The ultimate goal of my research is to transform humans - or at least himself - to AIs. This represents a pathway for humanity to transcend its biological limitations (e.g. immortality) and is the only way for modern humans to evolve. In addition to my academic pursuits, Weikang actively contributes to several open-source projects, such as Blender.
🏠 Homepage
Ngoc Bui
My research interests are machine learning and deep learning with a focus on their responsible use in real-world systems where humans and high measurement uncertainties exist in the loop. Recently, I'm focusing on multimodal large language models and their emerging capabilities in real-world applications.
🏠 Homepage
Hiren Madhu
Ever wonder what secrets lurk within the messy data most of the world holds? Graphs offer a way to untangle this mess, but the key lies in extracting knowledge from this organized chaos. That is where I come in. I'm particularly interested in answering how we can use machine learning methods with limited labeled data to learn representations of these geometric structures (i.e., graphs, simplicial complexes, etc.) and draw insights from them. Imagine unlocking valuable insights without needing mountains of hand-labeled data – the potential excites me! I want to push the boundaries of machine learning with limited labeled data. You might find me playing video games when I'm not tackling these data tangles. Assassin's Creed is my favorite game series.
🏠 Homepage
Yangtian Zhang
My research interests are focused on Generative Models, Graph Algorithms, and, more recently, Multi-Modal Foundation Models. Currently, I am exploring innovative solutions for real-world applications and scientific challenges.
🏠 Homepage
Siyi Gu
I am a first year Ph.D student and my primary research focus is Generative AI, particually post-training of LLMs and Multi-Modal Foundation Models. I am committed to developing AI technologies motivated by real-world application and societal benefits.
🏠 Homepage
Peiwen Li
My research focuses on LLM agents, with an emphasis on improving long-horizon reasoning and decision-making through reinforcement learning. I am particularly interested in building structured agentic frameworks that enable effective planning, collaboration, and specialization for complex real-world tasks. I also have experience in AutoML, OOD generalization, causal discovery, and graph learning. More broadly, I aim to develop scalable and reliable AI systems capable of reasoning over complex, structured, real-world data.
🏠 Homepage