Language models have steadily increased in size over the past few years....
The DialPort project http://dialport.org/, funded by the National Scienc...
The ultimate goal of dialog research is to develop systems that can be
e...
To facilitate zero-shot generalization in taskoriented dialog, this pape...
Instruction tuning is an emergent paradigm in NLP wherein natural langua...
This is a report on the NSF Future Directions Workshop on Automatic
Eval...
Developing mechanisms that flexibly adapt dialog systems to unseen tasks...
In transfer learning, it is imperative to achieve strong alignment betwe...
Automatic evaluation metrics are a crucial component of dialog systems
r...
This paper introduces the Ninth Dialog System Technology Challenge (DSTC...
While neural models have been shown to exhibit strong performance on
sin...
We present STAR, a schema-guided task-oriented dialog dataset consisting...
A key challenge of dialog systems research is to effectively and efficie...
A long-standing goal of task-oriented dialogue research is the ability t...
It is important to define meaningful and interpretable automatic evaluat...
The lack of meaningful automatic evaluation metrics for dialog has imped...
This paper discusses the importance of uncovering uncertainty in end-to-...
Voice-based technologies are typically developed for the average user, a...
Neural models of dialog rely on generalized latent representations of
la...
The aim of this paper is to mitigate the shortcomings of automatic evalu...
Neural dialog models have exhibited strong performance, however their
en...
This paper examines various unsupervised pretraining objectives for lear...
This paper makes the case that while most research on intelligent agents...
Despite being virtually ubiquitous, sequence-to-sequence models are
chal...