Shengchao Hu

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

The State of Reinforcement Finetuning for Transformer-based Agents

SeWA: Selective Weight Average via Probabilistic Masking

Rethinking the Role of Dynamic Sparse Training for Scalable Deep Reinforcement Learning

Rethinking Data Curation in LLM Training: Online Reweighting Offers Better Generalization than Offline Methods

Q-learning Penalized Transformer for Safe Offline Reinforcement Learning

Combatting Dimensional Collapse in LLM Pre-Training Data via Submodular File Selection

Analytic Energy-Guided Policy Optimization for Offline Reinforcement Learning

Tackling Continual Offline RL through Selective Weights Activation on Aligned Spaces

Continual Task Learning through Adaptive Policy Self-Composition

Solving Continual Offline RL through Selective Weights Activation on Aligned Spaces