Apologies for cross-posting.
Dear Colleagues and Students,
INFORMS Service Science Online Forum Series - Episode 8 features our next speaker Prof. Xi Chen (NYU Stern) who will share how cutting-edge stochastic optimization methods like ComPO and spectral policy optimization push large language models to align better with human preferences, stabilize training, and boost reasoning performance across model scales and benchmarks.
See you online for Episode 8 via this Zoom link, where AI meets service science!
Speaker: Prof. Xi Chen, NYU Stern
Moderator: Prof. Weiwei Chen, Rutgers
Title: LLM Alignment Techniques: Stochastic Optimizations in LLM Post-training and Reasoning
Abstract:
This talk explores approaches to improving large language model (LLM) post-training and reasoning through stochastic optimization techniques. The first part introduces ComPO, a preference alignment method using comparison oracles in stochastic optimization. The work addresses likelihood displacement issues in traditional direct preference optimization. The second part proposes the spectral policy optimization, a framework that overcomes GRPO's limitations with all-negative-sample groups by introducing response diversity with AI feedback. Both approaches demonstrate significant improvements across various model sizes and benchmarks, representing important advances in LLM post-training via stochastic optimization. This is a joint work with Peter Chen, Xiaopeng Li, Ziniu Li, Wotao Yin, and Tianyi Lin.
Time: December 15, 2025, Monday, 10:00 AM-11:00 AM EST
Zoom link: https://rutgers.zoom.us/j/95804877123?pwd=bipsDDJRbaULzgkgTBA0Wa1NaiaPhQ.1
Meeting ID: 958 0487 7123
Passcode: 095884
For more information, see our website https://sites.google.com/view/service-science-online-forum/, and our YouTube Channel https://www.youtube.com/playlist?list=PLCn8oCTLj5JEeIiA3_ATZp8gtlkWJCRpO.
------------------------------
Renyu Zhang
Associate Professor
The Chinese University of Hong Kong
Hong Kong
------------------------------