一作NeurIPS 2026Poster
Risk-Aware Action Repetition via Expected Skip Evaluation
Hyunwoo Park · Junhyeok Um · Baek-Ryun Seong · Sang-Ki Ko
评测与基准OpenReview
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。
Hyunwoo Park · Junhyeok Um · Baek-Ryun Seong · Sang-Ki Ko
Soohan Lim · Joonghyuk Hahn · Hyunwoo Park · Sang-Ki Ko · Yo-Sub Han
Current code generation evaluation measures functional correctness on well-formed inputs that satisfy all input preconditions. This paradigm has a critical limitation: task descriptions often leave these preconditions implicit, while evaluation filters out inputs that violate them. As a result, g...