⚠️ Unsupported Browser

Your browser is not supported.

The latest version of Safari, Chrome, Firefox, Internet Explorer or Microsoft Edge is required to use this website.

Click the button below to update and we look forward to seeing you soon.

Update now

Research

Academic Papers

ConVEx: Data-Efficient and Few-Shot Slot Labeling | In ArXiv e-prints, 2020 | Matthew Henderson, Ivan Vulić. 2020

 

ConveRT: Efficient and accurate conversational representations from transformers | Findings of EMNLP 2020 | Matthew Henderson, Inigo Casanueva, Nikola Mrkšić, Pei-Hao Su, Tsung-Hsein Wen, Ivan Vulić

 

Span-ConveRT: Few-shot Span Extraction for Dialog with Pretrained Conversational Representations | In Proceedings of ACL, pages 107-121 | Sam Coope, Tyler Farghly, Daniela Gerz, Ivan Vulić, and Matthew Henderson. 2020 | Data available on Github

 

Efficient Intent Detection with Dual Sentence Encoders | In Proceedings of the Workshop on NLP for Conversational AI, pages 38-45 | Iñigo Casanueva, Tadas Temčinas, Daniela Gerz, Matthew Henderson, and Ivan Vulić. 2020 | Data and code available on Github

 

PolyResponse: A Rank-based Approach to Task-Oriented Dialogue with Application in Restaurant Search and Booking | In Proceedings of EMNLP-IJCNLP: System Demonstrations, pages 181-186 | Matthew Henderson, Ivan Vulić, Iñigo Casanueva, Paweł Budzianowski, Daniela Gerz, Sam Coope, Georgios Spithourakis, Tsung-Hsien Wen, Nikola Mrkšić, and Pei-Hao Su. 2019

 

Hello, It’s GPT-2 – How Can I Help You? Towards the Use of Pretrained Language Models for Task-Oriented Dialogue Systems | In Proceedings of the 3rd Workshop on Neural Generation and Translation (WNGT@EMNLP 2019), pages 15-22 | Paweł Budzianowski and Ivan Vulić. 2019

 

JW300: A wide-coverage parallel corpus for low-resource languages | In Proceedings of ACL, pages 3204-3210 | Željko Agić and Ivan Vulić. 2019 | Data available at: http://opus.nlpl.eu/JW300.php

 

Training neural response selection for task-oriented dialogue systems | In Proceedings of ACL, pages 5392-5404. | Matthew Henderson, Ivan Vulić, Daniela Gerz, Iñigo Casanueva, Paweł Budzianowski, Sam Coope, Georgios Spithourakis, Tsung-Hsien Wen, Nikola Mrkšić, and Pei-Hao Su. 2019 | This model has been superseded by the more powerful pretraining method: ConveRT

 

A repository of conversational datasets | In Proceedings of the Workshop on NLP for Conversational AI, pages 1-10 | Matthew Henderson, Paweł Budzianowski, Iñigo Casanueva, Sam Coope, Daniela Gerz, Girish Kumar, Nikola Mrkšić, Georgios Spithourakis, Pei-Hao Su, Ivan Vulić, and Tsung-Hsien Wen. 2019 | Data available on Github

 

Fully statistical neural belief tracking | In Proceedings of ACL, pages 108-113 | Nikola Mrkšić and Ivan Vulić. 2018 

 

Reward estimation for dialogue policy optimisationIn Computer Speech and Language | Su et al. 2018 

 

Feudal Reinforcement Learning for Dialogue Management in Large Domains | NAACL  2018Casanueva et al. 2018 

 

Feudal Dialogue Management with Jointly Learned Feature Extractors | SIGDIAL 2018. | Casanueva et al. 2018 

 

MultiWOZ – A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modelling | EMNLP 2018 (Best Resource Paper) | Budzianowski et al. 2018 

 

Efficient Natural Language Response Suggestion for Smart ReplyIn ArXiv e-prints, 2017 | Matthew Henderson et al. 2017 

 

Semantic Specialisation of Distributional Word Vector Spaces using Monolingual and Cross-Lingual Constraints | In TACL | Mrkšić et al. 2017 

 

Neural Belief Tracker: Data-Driven Dialogue State TrackingACL  2017 Mrkšić et al. 2017 

 

A network-based end-to-end trainable task-oriented dialogue system | EACL 2017 | Wen et al. 2017 

 

Latent intention dialogue models | ICML 2017 | Wen et al. 2017 

 

Sample-efficient actor-critic reinforcement learning with supervised data for dialogue managementSIGDIAL 2017 | Su et al. 2017 

 

Counter-fitting word vectors to linguistic constraintsNAACL 2016  | Mrkšić et al. 2016 

 

On-line active reward learning for policy optimisation in spoken dialogue systemsACL 2016 (Best Student Paper) | Su et al. 2016 

 

A Benchmarking Environment for Reinforcement Learning Based Task Oriented Dialogue Management. Deep Reinforcement Learning Symposium | NIPS 2017 | Casanueva and Budzianowski et al. 2017 

 

Conditional generation and snapshot learning in neural dialogue systems | EMNLP 2016 | Wen et al. 2016

 

Knowledge transfer between speakers for personalised dialogue management | SIGDIAL 2015 | (Best Paper Nomination) | Casanueva et al. 2015 

 

Adaptive speech recognition and dialogue management for users with speech disorders | INTERSPEECH 2015 | Casanueva et al. 2015 

 

Semantically conditioned lstm-based natural language generation for spoken dialogue systems | EMNLP 2015 (Best Paper Award) | Wen et al. 2015 

 

Stochastic language generation in dialogue using recurrent neural networks with convolutional sentence reranking | SIGDIAL 2015 (Best Paper Award) | Wen et al. 2015 

 

Learning from real users: Rating dialogue success with neural networks for reinforcement learning in spoken dialogue systems | INTERSPEECH | Su et al. 2015 

 

The Second Dialog State Tracking ChallengeIn SIGDIAL | Matthew Henderson et al. 2014 

 

Word-based Dialog State Tracking with Recurrent Neural Networks |In SIGDIAL Matthew Henderson et al. 2014 

Get in Touch

Learn more about voice-based conversational AI, request a demo or find out how PolyAI can help.