模型:
microsoft/DialogRPT-human-vs-machine
请尝试这个 ➤➤➤ Colab Notebook Demo (click me!)
Context | Response | human_vs_machine score |
---|---|---|
I love NLP! | I'm not sure if it's a good idea. | 0.000 |
I love NLP! | Me too! | 0.605 |
"人机对话评分"是预测响应来自人类而不是机器的可能性的分数。
一个对话回复有多大可能会得到点赞?和/或回复?的评分?
这是通过训练于超过1亿人类反馈数据的一系列对话回复排名模型所预测的. 可以通过重新排名生成的回复候选项来改进现有的对话生成模型(例如 DialoGPT )。
快速链接:
我们考虑了以下任务,并提供了相应的预训练模型。
Task | Description | Pretrained model |
---|---|---|
Human feedback | given a context and its two human responses, predict... | |
updown | ... which gets more upvotes? | 1238321 |
width | ... which gets more direct replies? | 1239321 |
depth | ... which gets longer follow-up thread? | 12310321 |
Human-like (human vs fake) | given a context and one human response, distinguish it with... | |
human_vs_rand | ... a random human response | 12311321 |
human_vs_machine | ... a machine generated response | this model |
请在 our repo 上创建一个问题
@inproceedings{gao2020dialogrpt, title={Dialogue Response RankingTraining with Large-Scale Human Feedback Data}, author={Xiang Gao and Yizhe Zhang and Michel Galley and Chris Brockett and Bill Dolan}, year={2020}, booktitle={EMNLP} }