test accuracy
Test accuracy measures the proportion of correct predictions made by a model on a separate test set, reflecting how well the model generalizes beyond the training data.
- AdaSTaR: Adaptive Data Sampling for Training Self-Taught Reasoners
- Efficient Representativeness-Aware Coreset Selection
- Exploring Landscapes for Better Minima along Valleys
- LILO: Learning to Reason at the Frontier of Learnability
- Optimizing Chain-of-Thought Reasoners via Gradient Variance Minimization in Rejection Sampling and RL
- The Dual Nature of Plasticity Loss in Deep Continual Learning: Dissection and Mitigation
- The Rich and the Simple: On the Implicit Bias of Adam and SGD