RobustAPI

RobustAPI is a research tool designed to evaluate the robustness and reliability of large language models (LLMs) in generating code responses to programming questions sourced from StackOverflow, specifically focusing on 24 Java APIs. It includes a dataset of 1208 coding questions and provides evaluative scripts to identify API misuse in generated answers, enabling researchers to assess LLM performance against real-world coding scenarios. Notable features include integration with multiple popular LLMs and a Java-based API checker for enhanced evaluation accuracy.