评估最近大型预训练模型中的社会偏见（英文版）

英文标题：Evaluation of Social Biases in Recent Large Pre-Trained Models中文摘要：研究旨在探讨最新的预训练模型中是否比旧模型更负责任地发展，以便降低固有的社会偏见以及探究三个最近的模型（ELECTRA，DeBERTa 和 DistilBERT）在减少偏见方面的总体趋势。实验比较这些模型与基准 BERT 的关联指标，并发现所有研究对象都存在有偏差，但完整指标相对于 BERT 有所提高。英文摘要：Large pre-trained language models are widely used in the community. Thesemodels are usually trained on unmoderated and unfiltered data from open sourceslike the Internet. Due to this, biases that we see in platforms online whichare a reflection of those in society are in