Tian Jin

Tian Jin

Research Affiliate

Tian Jin is a 4th year PhD student at MIT CSAIL, working with Michael Carbin and Jonathan Ragan-Kelley. Tian's research focuses on sparsity -- the surprising phenomenon that a substantial fraction of weights in a neural network may be removed without hurting its predictive power. He uses empirical methods to dissect the mystery of sparsity from vision models to large language models.

Previously, he was a Research Engineer at IBM Thomas J. Watson Research Center, serving as the technical lead for enabling deep neural network model inference on IBM mainframe machines. He also contributed to the compiler support for IBM Summit Supercomputer, the most powerful supercomputer in the world in 2018.

Tian is a remote affiliate of the MINT lab since 2023, where he works with Seth Lazar and members of the MINT lab to develop large language model powered recommender systems that align with user values.