Pararth Shah
Pararth Shah has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Anuj Kumar
Facebook Conversational AI 1 Hacker Way 94025 Menlo Park CA USA
1 shared publication - Anusha Balakrishnan
Carnegie Mellon University Facebook AI
1 shared publication Caleb Bryant
Facebook Conversational AI Menlo Park CA
1 shared publicationDilek Hakkani-Tur
Amazon Alexa AI Mountain View CA USA
1 shared publication- Dongyeop Kang
School of Computer Science Carnegie Mellon University Pittsburgh PA USA
1 shared publication - Gokhan Tur
Google
1 shared publication
Latest Publications
- Open Access
- 1 January 2020
- Published by Association for Computational Linguistics (ACL)
- p. 290-296
Multi-Action Dialog Policy Learning with Interactive Human Teaching
Megha Jhunjhunwala, Caleb Bryant, Pararth Shah. Proceedings of the 21th Annual Meeting of the Special Interest Group on Discourse and Dialogue. 2020.
- 1 January 2020
- Published by Association for Computational Linguistics (ACL)
- p. 290-296
Open AccessMulti-Action Dialog Policy Learning with Interactive Human Teaching
Megha Jhunjhunwala, Caleb Bryant, Pararth Shah. Proceedings of the 21th Annual Meeting of the Special Interest Group on Discourse and Dialogue. 2020.
- Open Access
- 1 January 2019
- Published by Association for Computational Linguistics (ACL)
- p. 145-150
Memory Grounded Conversational Reasoning
Seungwhan Moon, Pararth Shah, Rajen Subba, Anuj Kumar. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations. 2019.
- 1 January 2019
- Published by Association for Computational Linguistics (ACL)
- p. 145-150
Open AccessMemory Grounded Conversational Reasoning
Seungwhan Moon, Pararth Shah, Rajen Subba, Anuj Kumar. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations. 2019.
- Open Access
- 1 January 2019
- Published by Association for Computational Linguistics (ACL)
- p. 1951-1961
Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue
Dongyeop Kang, Anusha Balakrishnan, Pararth Shah, Paul Crook, Y-Lan Boureau, Jason Weston. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.
- 1 January 2019
- Published by Association for Computational Linguistics (ACL)
- p. 1951-1961
Open AccessRecommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue
Dongyeop Kang, Anusha Balakrishnan, Pararth Shah, Paul Crook, Y-Lan Boureau, Jason Weston. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.
- Open Access
User Modeling for Task Oriented Dialogues
We introduce end-to-end neural network based models for simulating users of task-oriented dialogue systems. User simulation in dialogue systems is crucial from two different perspectives: (i) automatic evaluation of different dialogue models, and (ii) training task-oriented dialogue systems. We design a hierarchical sequence-to-sequence model that first encodes the initial user goal and system turns into fixed length representations using Recurrent Neural Networks (RNN). It then encodes the dialogue history using another RNN layer. At each turn, user responses are decoded from the hidden representations of the dialogue level RNN. This hierarchical user simulator (HUS) approach allows the model to capture undiscovered parts of the user goal without the need of an explicit dialogue state tracking. We further develop several variants by utilizing a latent variable model to inject random variations into user responses to promote diversity in simulated user responses and a novel goal regularization mechanism to penalize divergence of user responses from the initial user goal. We evaluate the proposed models on movie ticket booking domain by systematically interacting each user simulator with various dialogue system policies trained with different objectives and users.
Open AccessUser Modeling for Task Oriented Dialogues
We introduce end-to-end neural network based models for simulating users of task-oriented dialogue systems. User simulation in dialogue systems is crucial from two different perspectives: (i) automatic evaluation of different dialogue models, and (ii) training task-oriented dialogue systems. We design a hierarchical sequence-to-sequence model that first encodes the initial user goal and system turns into fixed length representations using Recurrent Neural Networks (RNN). It then encodes the dialogue history using another RNN layer. At each turn, user responses are decoded from the hidden representations of the dialogue level RNN. This hierarchical user simulator (HUS) approach allows the model to capture undiscovered parts of the user goal without the need of an explicit dialogue state tracking. We further develop several variants by utilizing a latent variable model to inject random variations into user responses to promote diversity in simulated user responses and a novel goal regularization mechanism to penalize divergence of user responses from the initial user goal. We evaluate the proposed models on movie ticket booking domain by systematically interacting each user simulator with various dialogue system policies trained with different objectives and users.