170 citations · 782 across the 19 of their papers we have counts for
25 papers
ConvFiT: Conversational Fine-Tuning of Pretrained Language Models
Ivan Vulić, Pei-Hao Su, Sam Coope +5
Transformer-based language models (LMs) pretrained on large text collections are proven to store a wealth of semantic knowledge. However, 1) they are not effective as sentence enco…
Multilingual and Cross-Lingual Intent Detection from Spoken Data
Daniela Gerz, Pei-Hao Su, Razvan Kusztos +6
We present a systematic study on multilingual and cross-lingual intent detection from spoken data. The study leverages a new resource put forth in this work, termed MInDS-14, a fir…
ConveRT: Efficient and Accurate Conversational Representations from Transformers
Matthew Henderson, Iñigo Casanueva, Nikola Mrkšić +3
General-purpose pretrained sentence encoders such as BERT are not ideal for real-world conversational AI applications; they are computationally heavy, slow, and expensive to train.…
PolyResponse: A Rank-based Approach to Task-Oriented Dialogue with Application in Restaurant Search and Booking
Matthew Henderson, Ivan Vulić, Iñigo Casanueva +7
We present PolyResponse, a conversational search engine that supports task-oriented dialogue. It is a retrieval-based approach that bypasses the complex multi-component design of t…
Training Neural Response Selection for Task-Oriented Dialogue Systems
Matthew Henderson, Ivan Vulić, Daniela Gerz +7
Despite their popularity in the chatbot literature, retrieval-based models have had modest impact on task-oriented dialogue systems, with the main obstacle to their application bei…
A Repository of Conversational Datasets
Matthew Henderson, Paweł Budzianowski, Iñigo Casanueva +8
Progress in Machine Learning is often driven by the availability of large datasets, and consistent evaluation metrics for comparing modeling approaches. To this end, we present a r…
Feudal Reinforcement Learning for Dialogue Management in Large Domains
Iñigo Casanueva, Paweł Budzianowski, Pei-Hao Su +4
Reinforcement learning (RL) is a promising approach to solve dialogue policy optimisation. Traditional RL algorithms, however, fail to scale to large domains due to the curse of di…
Sample Efficient Deep Reinforcement Learning for Dialogue Systems with Large Action Spaces
Gellért Weisz, Paweł Budzianowski, Pei-Hao Su +1
In spoken dialogue systems, we aim to deploy artificial intelligence to build automated dialogue agents that can converse with humans. A part of this effort is the policy optimisat…
A Benchmarking Environment for Reinforcement Learning Based Task Oriented Dialogue Management
Iñigo Casanueva, Paweł Budzianowski, Pei-Hao Su +6
Dialogue assistants are rapidly becoming an indispensable daily aid. To avoid the significant effort needed to hand-craft the required dialogue flow, the Dialogue Management (DM) m…
Reward-Balancing for Statistical Spoken Dialogue Systems using Multi-objective Reinforcement Learning
Stefan Ultes, Paweł Budzianowski, Iñigo Casanueva +6
Reinforcement learning is widely used for dialogue policy optimization where the reward function often consists of more than one component, e.g., the dialogue success and the dialo…
Sub-domain Modelling for Dialogue Management with Hierarchical Reinforcement Learning
Paweł Budzianowski, Stefan Ultes, Pei-Hao Su +5
Human conversation is inherently complex, often spanning many different topics/domains. This makes policy learning for dialogue systems very challenging. Standard flat reinforcemen…
Sample-efficient Actor-Critic Reinforcement Learning with Supervised Data for Dialogue Management
Pei-Hao Su, Pawel Budzianowski, Stefan Ultes +2
Deep reinforcement learning (RL) methods have significant potential for dialogue policy optimisation. However, they suffer from a poor performance in the early stages of learning.…
Exploiting Sentence and Context Representations in Deep Neural Models for Spoken Language Understanding
Lina M. Rojas Barahona, Milica Gasic, Nikola Mrkšić +4
This paper presents a deep learning architecture for the semantic decoder component of a Statistical Spoken Dialogue System. In a slot-filling dialogue, the semantic decoder predic…
Dialogue manager domain adaptation using Gaussian process reinforcement learning
Milica Gasic, Nikola Mrksic, Lina M. Rojas-Barahona +5
Spoken dialogue systems allow humans to interact with machines using natural speech. As such, they have many benefits. By using speech as the primary communication medium, a comput…
Conditional Generation and Snapshot Learning in Neural Dialogue Systems
Tsung-Hsien Wen, Milica Gasic, Nikola Mrksic +5
Recently a variety of LSTM-based conditional language models (LM) have been applied across a range of language generation tasks. In this work we study various model architectures a…
Continuously Learning Neural Dialogue Management
Pei-Hao Su, Milica Gasic, Nikola Mrksic +5
We describe a two-step approach for dialogue management in task-oriented spoken dialogue systems. A unified neural network framework is proposed to enable the system to first learn…
On-line Active Reward Learning for Policy Optimisation in Spoken Dialogue Systems
Pei-Hao Su, Milica Gasic, Nikola Mrksic +5
The ability to compute an accurate reward function is essential for optimising a dialogue policy via reinforcement learning. In real-world applications, using explicit user feedbac…
A Network-based End-to-End Trainable Task-oriented Dialogue System
Tsung-Hsien Wen, David Vandyke, Nikola Mrksic +5
Teaching machines to accomplish tasks by conversing naturally with humans is challenging. Currently, developing task-oriented dialogue systems requires creating multiple components…
Multi-domain Neural Network Language Generation for Spoken Dialogue Systems
Tsung-Hsien Wen, Milica Gasic, Nikola Mrksic +4
Moving from limited-domain natural language generation (NLG) to open domain is difficult because the number of semantic input combinations grows exponentially with the number of do…
Counter-fitting Word Vectors to Linguistic Constraints
Nikola Mrkšić, Diarmuid Ó Séaghdha, Blaise Thomson +6
In this work, we present a novel counter-fitting method which injects antonymy and synonymy constraints into vector space representations in order to improve the vectors' capabilit…
Semantically Conditioned LSTM-based Natural Language Generation for Spoken Dialogue Systems
Tsung-Hsien Wen, Milica Gasic, Nikola Mrksic +3
Natural language generation (NLG) is a critical component of spoken dialogue and it has a significant impact both on usability and perceived quality. Most NLG systems in common use…
Reward Shaping with Recurrent Neural Networks for Speeding up On-Line Policy Learning in Spoken Dialogue Systems
Pei-Hao Su, David Vandyke, Milica Gasic +3
Statistical spoken dialogue systems have the attractive property of being able to be optimised from data via interactions with real users. However in the reinforcement learning par…
Learning from Real Users: Rating Dialogue Success with Neural Networks for Reinforcement Learning in Spoken Dialogue Systems
Pei-Hao Su, David Vandyke, Milica Gasic +4
To train a statistical spoken dialogue system (SDS) it is essential that an accurate method for measuring task success is available. To date training has relied on presenting a tas…
Stochastic Language Generation in Dialogue using Recurrent Neural Networks with Convolutional Sentence Reranking
Tsung-Hsien Wen, Milica Gasic, Dongho Kim +4
The natural language generation (NLG) component of a spoken dialogue system (SDS) usually needs a substantial amount of handcrafting or a well-labeled dataset to be trained on. The…