SEGAgentRL

AI & ML interests

We target improved agent reinforcement learning in terms of stability (S), efficiency (E), and generalization (G).

Recent Activity

dwenlong  submitted a paper about 1 month ago

Learning Dynamics of Continual Learning: A Unified View of Data Attribution, Forgetting, and Plasticity Loss

dwenlong  submitted a paper 5 months ago

Directional Alignment Mitigates Reward Hacking in Reinforcement Learning for Language Models

dwenlong  submitted a paper 6 months ago

For-Value: Efficient Forward-Only Data Valuation for finetuning LLMs and VLMs

View all activity

Team members 2