DART: Distributional Adversarial Recurrent Training for Algorithm Learning
Recurrent reasoning models (RRMs) can solve structured problems, achieving easy-to-hard generalization through iterative computation in hidden space. These models are typically trained with instance-level supervision, which becomes increasingly problematic as task difficulty grows: valid solutions occupy a tiny region...
Hieu Tran Bao, Phung Thanh Dang, Minh Quang Nhat Pham et al.
· 0 citations