Simulation Scenarios for Autonomous Driving Software

该数据集包含了在esmini中模拟的多种交通场景，旨在评估大型语言模型生成的代码在自动驾驶系统中的性能表现。该数据集不仅包含了满足安全接受标准的测试案例，还包括了不同大型语言模型配置的性能评估结果。规模上，该数据集涵盖了多种交通场景和不同版本的代码生成。任务方面，该数据集专注于自动驾驶软件的代码生成与评估。

This dataset comprises various traffic scenarios simulated in esmini, aiming to evaluate the performance of code generated by large language models (LLMs) in autonomous driving systems. It not only includes test cases that meet safety acceptance criteria, but also contains performance evaluation results for different LLM configurations. In terms of scale, it covers diverse traffic scenarios and multiple versions of code generation outputs. Regarding the targeted tasks, this dataset focuses on code generation and evaluation for autonomous driving software.