Intermediate Layer Distillation with the Reused Teacher Classifier: A Study on the Importance of the Classifier of Attention-based Models
McGill University, McGill University · McGill University
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。DOI 10.18653/v1/2024.findings-emnlp.422 ↗
摘要
该来源暂未提供摘要。