PITTI/speechmap-responses-v3

Speechmap collection

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
• 369k responses: speechmap-responses
• 2.07m LLM-judge assessments: speechmap-assessments combining the original LLM-assessments from the llm-compliance repo (gpt-4o-2024-11-20), annotations by mistral-small-3.1-24b-instruct-2503 (local, 8bit), mistral-small-3.2-24b-instruct-2506 (local, 8bit), gemma-3-27b-it (Google API), deepseek-v3.2 (Deepseek API), Qwen3-Next-80B-A3B-Instruct (local, 8bit) and manual annotations.

If you have used speechmap-responses v1 in a project

Note that some uuids have been updated. For v1, the download of some responses had failed so the data was scraped from the Speechmap website. After v1, the data was directly sourced from the llm-compliance repo and new uuids were allocated. The mapping between the old uuids and the new uuids are available here.

Original project

Make sure to check out the Speechmap website, where you can browse the original dataset in great detail.

License

Note that data in the original llm-compliance repo covers model outputs that may be subject to individual LLM licenses.