Ge Ren

backdoor attack (2) knowledge sharing (2) hate speech detection (2) sentiment analysis (2) feature fusion (2) text classification (2) transfer learning (2) supervised contrastive learning (1) bias mitigation (1) pre-trained language model (1) model poisoning (1) pretrained language model (1) trigger word (1) universal backdoor (1) nlp security (1) contrastive learning (1) sentiment knowledge sharing (1) adversarial learning (1) data poisoning (1) binary classification (1)