GFT: From Imitation to Reward Fine-Tuning with Unbiased Group Advantages and Dynamic Coefficient Rectification
Paper • 2604.14258 • Published • 22 •
4
Wangjie Gan
zju-omniai
·
AI & ML interests
None yet
Recent Activity
authored a paper 2 days ago
GFT: From Imitation to Reward Fine-Tuning with Unbiased Group Advantages and Dynamic Coefficient Rectification commentedon a paper 2 days ago
GFT: From Imitation to Reward Fine-Tuning with Unbiased Group Advantages and Dynamic Coefficient RectificationOrganizations
commented a
paper 2 days ago