Details Abstract: Multiple choice questions (MCQs) serve as a common yet important task format in the research of large language models (LLMs). This work shows that LLMs are vulnerable to option ...
一些您可能无法访问的结果已被隐去。
显示无法访问的结果一些您可能无法访问的结果已被隐去。
显示无法访问的结果