作者位次 #4/8NeurIPS 2026Poster
Can LLMs Reliably Grade Olympiad Proofs? A Controlled Study of Mathematical Verification with LLMs
Azim Ospanov · Zijin Feng · Ding Ding · Chengwu Liu · Haoli Bai · Jiacheng Sun · Lifeng Shang · Farzan Farnia
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Azim Ospanov · Zijin Feng · Ding Ding · Chengwu Liu · Haoli Bai · Jiacheng Sun · Lifeng Shang · Farzan Farnia
Chengwu Liu · Yichun Yin · Ye Yuan · Jiaxuan Xie · Botao Li · Siqi Li · Jianhao Shen · Yan Xu · Lifeng Shang · Ming Zhang
Most ATP benchmarks embed the final answer within the formal statement — a convention we call “Easy Mode” — a design that simplifies the task relative to what human competitors face and may lead to optimistic estimates of model capability.We call the stricter, more realistic setting “Hard Mode”:...