György Szarvas

multi-task learning (1) few-shot learning (1) sentiment analysis (1) text classification (1) model calibration (1) preference learning (1) knowledge distillation (1) cross-lingual transfer (1) task-oriented dialogue (1) neural architecture (1) class imbalance (1) out-of-distribution detection (1) semantic clustering (1) uncertainty estimation (1) zero-shot transfer (1) cross-entropy loss (1) focal loss (1) pretrained language model (1) hierarchical sampling (1) zero-shot learning (1)