This project studies Data-centric Solution Preference—predicting which ML solution will perform better before executing it—by leveraging data analysis context and LLM reasoning. The repository provides curated solution corpora, task resources, agent-run outputs, and analysis artifacts to support both the main evaluation… See the full description on the dataset page:
https://huggingface.co/datasets/zjunlp/PredictBeforeExecute.
