Natural Language Inference(NLI) is the task of predicting the labels(entailment, contradiction, and neutral,) for sentence pairs.
People invent a lot of deep model to solve this problem.
But I can not think of some application or scenario for this kind of deep model.
Dear Tong Guo,
Look the link, maybe useful.
https://www.intercom.com/customizable-bots?utm_source=google&utm_medium=sem&utm_campaign=1523805115&utm_term=%2Bchat%20%2Bbot&utm_ad_collection=59812237673&_bt=339017868413&_bg=59812237673&utm_ad=339017868413&offer=sponsoredby&utm_campaign_name=go_evg_acq_trial_nb-b_chatbot_bld_row_en&utm_ad_collection_name=sup-p_chat-bot-bmm&utm_ad_name=sponsoredby_text_20q2&gclid=CjwKCAjwyqTqBRAyEiwA8K_4O9Gb9N9zSPulceFR2w13dC92XN5oJ6QsOwywKjq8CfAW4FQBCAsaWRoCZlMQAvD_BwE
https://www.accenture.com/t00010101t000000__w__/br-pt/_acnmedia/pdf-45/accenture-chatbots-customer-service.pdf
Regards,
Shafagat
I'd suggest you have a look at following three papers to grasp how NLI is used in dialog processing.
https://www.aclweb.org/anthology/P19-1363 https://arxiv.org/abs/1903.05759 https://arxiv.org/abs/1905.05704
Best,
Rafal
I am not aware about how they are being used in chatbots but there are other places where they have found their applications. Here are some papers regarding application of entailment :
1. http://kozareva.com/papers/nldb06mlent.pdf
2. http://www.cfilt.iitb.ac.in/resources/surveys/te-mt_shubham-june14.pdf
- Badges
- Science topic