Jinyang Li

Professor of Computer Science

Department of Computer Science
Courant Institute of Mathematical Sciences
New York University

60 5th Ave, New York, NY

Research Interests

Machine learning systems, distributed and parallel computing, fault tolerant systems

Selected Open Source Projects

Grendel: Distributed 3D Gaussian Splatting

A distributed training system that enables unprecedented scaling of Guassian Splatting and other point-based rendering algorithms for 3D reconstruction.

Github ICLR'25 Oral

A fast and memory-efficient message-passing programming framework for training Graph Neural Networks.

Github Paper

Recent Publications All

[OSDI 2025] Understanding Stragglers in Large Model Training Using What-if Analysis
[ICLR 2025] On Scaling Up 3D Gaussian Splatting Training
[NSDI 2025] Runtime Protocol Refinement Checking for Distributed Protocol Implementations
[SIGMOD 2025] Automated Validating and Fixing of Text-to-SQL Translation with Execution Consistency
[SIGMOD 2025] DiskGNN: Bridging I/O Efficiency and Model Accuracy for Out-of-Core GNN Training
[Eurosys 2025] Stateful Large Language Model Serving with Pensiev

Teaching

[Fall 2025] Machine Learning Systems
[Spring 2026] Computer Systems Organization

Students

I am co-advising many wonderful Ph.D. students with Prof. Panda

Ding Ding
Haitian Jiang
Jinkun Lin
David Pissara
Anqi Zhang
Hexu Zhao