Jiaxiang Li

Memory-Efficient LLM Pretraining via Minimalist Optimizer Design

A Tale of Two Problems: Multi-Objective Bilevel Learning Meets Equality Constrained Multi-Objective Optimization

Muon Outperforms Adam in Tail-End Associative Memory Learning

Aligning Frozen LLMs by Reinforcement Learning: An Iterative Reweight-then-Optimize Approach

ADARL: Adaptive Low-Rank Structures for Robust Policy Learning under Uncertainty

Joint Reward and Policy Learning with Demonstrations and Human Feedback Improves Alignment

Policy optimization can be memory-efficient: LLM Alignment Through Successive Policy Re-weighting (SPR)