Zexue He

bias mitigation (4) large language model (4) text generation (3) contrastive learning (2) text rewriting (2) data augmentation (2) sensitive attribute (2) natural language processing (2) interpretable prediction (2) zero-shot learning (2) sentiment analysis (1) weakly supervised learning (1) knowledge distillation (1) preference alignment (1) attention mechanism (1) domain adaptation (1) self-supervised learning (1) cross-lingual transfer (1) feature selection (1) transfer learning (1)