Yuyang Ding

reinforcement learning (1) zero-shot learning (1) semi-supervised learning (1) text classification (1) mathematical reasoning (1) preference alignment (1) question answering (1) preference optimization (1) test-time adaptation (1) cross-modal transfer (1) instruction tuning (1) label noise (1) distribution shift (1) gaussian mixture model (1) language model (1) credit assignment (1) data synthesis (1) pretrained language model (1) mixup training (1) text detoxification (1)