Peng Wang

Understanding the Dynamics of Forgetting and Generalization in Continual Learning via the Neural Tangent Kernel

Theoretical Guarantees for Iterative Alignment of Self-Rewarding Language Models

MergOPT: A Merge-Aware Optimizer for Robust Model Merging

The State of Reinforcement Finetuning for Transformer-based Agents

SeWA: Selective Weight Average via Probabilistic Masking

Q-learning Penalized Transformer for Safe Offline Reinforcement Learning

Towards Understanding Memory buffer based Continual Learning