Sijin Chen — Princeton University

About

I am a third-year Ph.D. candidate in the Department of Electrical and Computer Engineering at Princeton University, advised by Prof. Jason M. Klusowski. Previously, I obtained a B.Sc. with First Class Honours in Computer Science from the Chinese University of Hong Kong, with a minor in Mathematics.

My research interests lie in scalable algorithms and statistical foundations for discrete generative models, including:

diffusion language models (DLMs), discrete diffusion, flow and consistency modeling for language;
statistical foundations of LLM decoding, policy-based inference, and connections to stochastic and robust optimization.

Outside research, I enjoy competitive soccer (football), tennis, and piano.

Publications

On Confidence-Based Decoding Strategies for Masked Diffusion and Beyond: A Likelihood Optimization Perspective.

S. Chen, C. Yu, J. Shi, J. M. Klusowski

Submitted to Conference on Neural Information Processing Systems (NeurIPS), 2026.
Multi-Mask Diffusion Language Models for Few-Step Generation.

S. Chen*, Y. Ren*, H. Zhao, Z. Cheng, Q. Gu, L. Ying

Conference on Language Modeling (COLM), 2026.
Beyond Masks: Efficient, Flexible Diffusion Language Models via Deletion-Insertion Processes.

F. Ding*, D. Ding*, S. Chen*, K. Wang, P. Xu, Z. Feng, H. Bai, K. Han, Y. Yan, B. Yuan, J. Sun

International Conference on Learning Representations (ICLR), 2026.

Post on X
Decoding Game: On Minimax Optimality of Heuristic Text Generation Strategies.

S. Chen, O. Hagrass, and J. M. Klusowski

International Conference on Learning Representations (ICLR), 2025.

Post on X
Escaping Saddle Points in Heterogeneous Federated Learning via Distributed SGD with Communication Compression.

S. Chen, Z. Li, and Y. Chi

International Conference on Artificial Intelligence and Statistics (AISTATS), 2024.
Non-Convex Joint Community Detection and Group Synchronization via Generalized Power Method.

S. Chen, X. Cheng, and A. M.-C. So

International Conference on Artificial Intelligence and Statistics (AISTATS), 2024.
CIA-SSD: Confident IoU-Aware Single-Stage Object Detector from Point Cloud.

W. Zheng, W. Tang, S. Chen, L. Jiang, and C.-W. Fu

AAAI Conference on Artificial Intelligence (AAAI), 2021.