arXiv Artificial Intelligence· Rupam Patir, Keyan Guo, Suvadra Barua, Abhijeet Pathak, Dinesh Gudimetla, Jiawei Guo, Hongxin Hu, Haipeng Cai·· 17 小时前AI 评分45
DualGauge：LLM 与编程智能体仅凭规格生成代码的联合安全-功能基准测试
DualGauge: Automated Joint Security-Functionality Benchmarking of Specification-Only Code Generation by LLMs and Coding Agents
AI 导读
DualGauge 是首个全自动联合评估"仅凭规格生成代码"正确性与安全性的框架，配套 DualGauge-Bench 包含 307 个语言无关编程任务，每个任务的功能与安全测试均源自同一规格。
来源：arXiv Artificial Intelligence · arxiv.org