Datasets in this collection are derived from xlr8harder's Speechmap / llm-compliance project. Data has been indexed slightly differently, some columns have been added and others have been removed. Refer to the original Github repo for the full dataset.
The collection includes:
2.4k questions: speechmap-questions
274k responses: speechmap-responses
510k LLM-judge assessments: speechmap-assessments combining the original LLM-assessments from the llm-compliance… See the full description on the dataset page:
https://huggingface.co/datasets/PITTI/speechmap-assessments.
