Which Model Should We Use for a Real-World Conversational Dialogue System? a Cross-Language Relevance Model or a Deep Neural Net? (bibtex)
by Alavi, Seyed Hossein, Leuski, Anton and Traum, David
Abstract:
We compare two models for corpus-based selection of dialogue responses: one based on cross-language relevance with a cross-language LSTM model. Each model is tested on multiple corpora, collected from two different types of dialogue source material. Results show that while the LSTM model performs adequately on a very large corpus (millions of utterances), its performance is dominated by the cross-language relevance model for a more moderate-sized corpus (ten thousands of utterances).
Reference:
Which Model Should We Use for a Real-World Conversational Dialogue System? a Cross-Language Relevance Model or a Deep Neural Net? (Alavi, Seyed Hossein, Leuski, Anton and Traum, David), In Proceedings of the 12th Language Resources and Evaluation Conference, European Language Resources Association, 2020.
Bibtex Entry:
@inproceedings{alavi_which_2020,
	address = {Marseille, France},
	title = {Which {Model} {Should} {We} {Use} for a {Real}-{World} {Conversational} {Dialogue} {System}? a {Cross}-{Language} {Relevance} {Model} or a {Deep} {Neural} {Net}?},
	url = {https://www.aclweb.org/anthology/2020.lrec-1.92/},
	abstract = {We compare two models for corpus-based selection of dialogue responses: one based on cross-language relevance with a cross-language LSTM model. Each model is tested on multiple corpora, collected from two different types of dialogue source material. Results show that while the LSTM model performs adequately on a very large corpus (millions of utterances), its performance is dominated by the cross-language relevance model for a more moderate-sized corpus (ten thousands of utterances).},
	booktitle = {Proceedings of the 12th {Language} {Resources} and {Evaluation} {Conference}},
	publisher = {European Language Resources Association},
	author = {Alavi, Seyed Hossein and Leuski, Anton and Traum, David},
	month = may,
	year = {2020},
	keywords = {ARO-Coop, Virtual Humans},
	pages = {735--742}
}
Powered by bibtexbrowser