Zhengmian Hu

Research Scientist/Engineer

San Jose

Zhengmian Hu is a researcher at Adobe Research. He received his Ph.D. in Computer Science from the University of Maryland, College Park, where he was advised by Professor Heng Huang. His research interests include optimization, deep learning theory, and probabilistic methods in machine learning. His work has been published at top machine learning venues such as ICML, NeurIPS, and ICLR.

Publications

A Personalized Conversational Benchmark: Towards Simulating Personalized Conversations

Li, Li., Cai, Peilin., Rossi, Ryan., Dernoncourt, Franck., Kveton, Branislav., Wu, Junda., Yu, Tong., Song, Linxin., Yang, Tiankai., Qin, Yuehan., Ahmed, Nesreen., Basu, Samyadeep., Mukherjee, Subhojyoti., Zhang, Ruiyi., Hu, Zhengmian., Ni, Bo., Zhou, Yuxiao., Wang, Zichao., Huang, Yue., Wang, Yu., Zhang, Xiangliang., Yu, Philip., Hu, Xiyang., Zhao, Yue. (Dec. 7, 2025)

Spotlight paper (top 5%)

NeurIPS 2025 MTI-LLM workshop

A Survey on Small Language Models

Van Nguyen, Chien., Shen, Xuan., Aponte, Ryan., Xia, Yu., Basu, Samyadeep., Hu, Zhengmian., Chen, Jian., Parmar, Mihir., Kunapuli, Sasidhar., Barrow, Joe., Wu, Junda., Singh, Ashish., Wang, Yu., Gu, Jiuxiang., Ahmed, Nesreen., Lipka, Nedim., Zhang, Ruiyi., Chen, Xiang., Yu, Tong., Kim, Sungchul., Deilamsalehy, Hanieh., Park, Namyong., Rimer, Michael., Zhang, Zhehao., Yang, Huanrui., Mathur, Puneet., Wu, Gang., Dernoncourt, Franck., Rossi, Ryan., Nguyen, Thien. (Sep. 10, 2025)

RANLP 2025