arXiv cs.SE· Yuanzhe Liu, Bastian Lipka, Avraham Shinnar, Jatin Ganhotra, Yao Ma, Martin Hirzel·· 4 小时前AI 评分38
研究指出编码基准任务描述过度泄露真值会削弱评测难度
Between Spoilers and Guesswork: Avoiding Over-Specification in Coding Benchmarks
AI 导读
arXiv 论文《Between Spoilers and Guesswork: Avoiding Over-Specification in Coding Benchmarks》指出。
来源：arXiv cs.SE · arxiv.org