作者位次 #5/6ICLR 2026Poster
Code2Bench: Scaling Source and Rigor for Dynamic Benchmark Construction
Zhe Zhang · Runlin Liu · Aishan Liu · Xingyu Liu · Xiang Gao · Hailong Sun
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。
Zhe Zhang · Runlin Liu · Aishan Liu · Xingyu Liu · Xiang Gao · Hailong Sun
Xiao Li · Runlin Liu · Zhe Zhang · Xiang Gao · Hailong Sun
Large Language Models (LLMs) have demonstrated remarkable proficiency in code generation, yet their application to Property-Based Testing (PBT) remains fraught with a superficiality gap. While LLMs can readily generate syntactically correct tests, they often struggle to bridge the semantic gap be...