Web API Invocation Tasks Dataset
收藏资源简介:
本数据集包含四个真实世界API的Web API调用任务,共近400对任务和预期结果。数据集用于执行基础上的正确性分析。数据集的内容包括API调用的HTTP方法、URL、请求体、请求头和查询参数。数据集的创建过程使用了大型语言模型(LLM)和API规范,并经过手动审查和修正。数据集的应用领域是评估LLMs生成Web API调用代码的能力,旨在解决API集成代码生成中的挑战。
This dataset contains Web API calling tasks from four real-world APIs, with nearly 400 pairs of tasks and their corresponding expected outcomes. It is used for correctness analysis based on the included task data. The dataset covers HTTP methods, URLs, request bodies, request headers, and query parameters involved in API calls. The dataset was created using large language models (LLMs) and API specifications, and underwent manual review and revision. Its purpose is to evaluate the ability of LLMs to generate Web API calling code, aiming to resolve the challenges encountered in API integration code generation.




