Fetching the paper…

Identifying and Measuring Token-Level Sentiment Bias in Pre-trained Language Models with Prompts · Around