Liangliang Shi

Bio

Dr. Liangliang Shi obtained his Ph.D. degree from the Department of Computer Science and Engineering at Shanghai Jiao Tong University under the supervision of Professor Junchi Yan from Rethinklab. He joined SIMIS as a tenure-track Assistant Professor in April 2025. His research interests primarily focus on entropic optimal transport, deep generative models, and combinatorial optimization. He has published over ten papers in top-tier CCF or CAAI A conferences such as NeurIPS, ICML, and ICLR. Additionally, he has served as a reviewer for top-tier conferences and journals, including NeurIPS, ICML, ICLR, KDD, AAAI, and TNNLS over the past several years.

Education Experience

• 2020.9-2025.3   Shanghai Jiao Tong University   Department of Computer Science and Engineering   Doctor

Work Experience

Honors and Awards

Publications

[1] Liangliang Shi, Yufeng Li, Kaipeng Zeng, Yihui Tu, Junchi Yan. Optimal  Flow  Transport  and  its  Entropic  Regularization:  a  GPU-friendly  Matrix  Iterative Algorithm for Flow Balance Satisfaction. In ICLR 2025.

[2] Liangliang Shi, Zhengyan Shi, Junchi Yan. SelKD: Selective Knowledge Distillation via Optimal Transport Perspective. In ICLR 2025.

[3] Liangliang Shi, Jack Fan, Junchi Yan. OT-CLIP: Understanding and Generalizing CLIP via Optimal Transport. In ICML 2024.

[4] Guolong Wang, Xun Wu, Zheng Qin, Liangliang Shi (corresponding author). Routing Evidence for Unseen Actions in Video Moment Retrieval. In KDD 2024.

[5] Liangliang Shi, Gu Zhang, Haoyu Zhen, Jintao Fan, Junchi Yan. Understanding and Generalizing Contrastive Learning from the Inverse Optimal Transport Perspective. In ICML 2023.

[6] Liangliang Shi, Zhaoqi Shen, Junchi Yan. Double-Bounded Optimal Transport for Advanced Clustering and Classification. In AAAI 2024.

[7] Liangliang Shi, Haoyu Zhen, Gu Zhang, Junchi Yan. Relative entropic optimal transport: a (prior-aware) matching perspective to (unbalanced) classification. In NeurIPS 2023.

[8] Yang Li, Liangliang Shi, Junchi Yan. IID-GAN: an IID Sampling Perspective for Regularizing Mode Collapse. In IJCAI 2023.

[9] Yang Li, Yichuan Mo, Liangliang Shi, Junchi Yan. Improving generative adversarial networks via adversarial learning in latent space. In NeurIPS 2022 (spotlight).

[10] Jianchao Zhu, Liangliang Shi, Junchi Yan, Hongyuan Zha. AutoMix: Mixup Networks for Sample Interpolation via Cooperative Barycenter Learning. In ECCV 2020.