1 paper · 1 filter
Lingyao Li, Junjie Xiong, Changjia Zhu +5
Large language models (LLMs) are increasingly used in academic peer review, yet their reliability, alignment with human judgment, and robustness to adversarial attacks remain poorl…