A modified dataset of English dialogs between a user and an assistant discussing movie preferences in natural language. In the foundational dataset, it initially had 502 records.
Dataset Information
Name: ccpemv2.jsonl
Version: Version 0.2
Modifications Included: Transformed to ### User:... ### Assistant:... format
Language: English
License: Creative Commons Attribution 4.0
Original Dataset Citation
@inproceedings{radlinski-etal-2019-ccpe,
title = {Coached Conversational Preference… See the full description on the dataset page:
https://huggingface.co/datasets/aloobun/ccpemv2.
