rishabhjoshi / DialoGraph_ICLR21

Repository for ICLR 2021 paper DialoGraph: Incorporating Interpretable Strategy-Graph Networks into Negotiation Dialogues

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

DialoGraph

Incorporating Interpretable Strategy-Graph Networks into Negotiation Dialogues


src/bot has the main model code.

src/preproc has the preprocessing code to extract strategies and then convert data to the format that the training script requires.

src/wfst has fst related code / data

src/yiheng_findfeatures has the code to extract strategies as written by Yiheng Zhou.

Download the processed data from here.

Citation:

Please cite the following paper if you use this code in your work.

@inproceedings{
joshi2021dialograph,
title={DialoGraph: Incorporating Interpretable Strategy-Graph Networks into Negotiation Dialogues},
author={Rishabh Joshi and Vidhisha Balachandran and Shikhar Vashishth and Alan Black and Yulia Tsvetkov},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=kDnal_bbb-E}
}

For any clarification, comments, or suggestions please create an issue or contact Rishabh.

About

Repository for ICLR 2021 paper DialoGraph: Incorporating Interpretable Strategy-Graph Networks into Negotiation Dialogues

License:Apache License 2.0


Languages

Language:HTML 71.1%Language:Python 14.7%Language:Jupyter Notebook 13.9%Language:CSS 0.2%Language:Shell 0.1%Language:JavaScript 0.1%Language:Dockerfile 0.0%