FlowDelta: Modeling Flow Information Gain in Reasoning for Conversational Machine Comprehension

An implementation of the FlowDelta: Modeling Flow Information Gain in Reasoning for Conversational Machine Comprehension

CodaLab pages submitted to public leaderboard:

QuAC

CoQA

As there are large differences between the usage of FlowDeltaQA and BertFlowDelta, we place the usage seperately in the corresponding directory.

Reference:

Main paper to be cited

@inproceedings{yeh2019flowdelta,
  title={FlowDelta: Modeling Flow Information Gain in Reasoning for Conversational Machine Comprehension},
  author={Yeh, Yi-Ting and Chen, Yun-Nung},
  booktitle={Proceedings of the 2nd Workshop on Machine Reading for Question Answering},
  pages={86--90},
  year={2019}
}