Recent research has shown that multi-task pre-training greatly improves ...
Most existing task-oriented dialog (TOD) systems track dialog states in ...
Energy-based language models (ELMs) parameterize an unnormalized distrib...
Maximum likelihood (ML) learning for energy-based models (EBMs) is
chall...
Building user simulators (USs) for reinforcement learning (RL) of
task-o...
Recently, there has been progress in supervised funetuning pretrained GP...
Recently, there have merged a class of task-oriented dialogue (TOD) data...
Developing semi-supervised task-oriented dialog (TOD) systems by leverag...
A challenge on Semi-Supervised and Reinforced Task-Oriented Dialog Syste...
Recently, Transformer based pretrained language models (PLMs), such as G...
Utilizing text-only data with an external language model (LM) in end-to-...
History and future contextual information are known to be important for
...
Recently, the end-to-end training approach for multi-channel ASR has sho...
Recovering programs' call graphs is crucial for inter-procedural analysi...
Recently, two approaches, fine-tuning large pre-trained language models ...
The use of phonological features (PFs) potentially allows language-speci...
Automatic speech recognition systems have been largely improved in the p...
Time Delay Neural Networks (TDNNs) are widely used in both DNN-HMM based...
Automatic speech recognition (ASR) has been significantly advanced with ...
Neural Architecture Search (NAS), the process of automating architecture...
A class of recent semi-supervised learning (SSL) methods heavily rely on...
Structured belief states are crucial for user goal tracking and database...
Although with progress in introducing auxiliary amortized inference mode...
In this paper, we present a new open source toolkit for speech recogniti...
Neural generative models have achieved promising performance on dialog
g...
There has been a long recognition that discrete features (n-gram feature...
Conversations have an intrinsic one-to-many property, which means that
m...
In this paper, we present a new open source toolkit for automatic speech...
Conditional random fields (CRFs) have been shown to be one of the most
s...
Slot filling is a crucial component in task-oriented dialog systems, whi...
This document aims to provide a review on learning with deep generative
...
In this paper, we present an end-to-end automatic speech recognition sys...
A new whole-sentence language model - neural trans-dimensional random fi...
In this paper we develop Neural Random Field learning with
Inclusive-div...
An ensemble of neural networks is known to be more robust and accurate t...
Dialog state tracking (DST) is a crucial component in a task-oriented di...
Trans-dimensional random field language models (TRF LMs) where sentences...
Trans-dimensional random field language models (TRF LMs) have recently b...
The dominant language models (LMs) such as n-gram and neural network (NN...
Though with progress, model learning and performing posterior inference ...
Existing MAP inference algorithms for determinantal point processes (DPP...