The Serverless Computing is becoming increasingly popular due to its eas...
Graph neural networks (GNNs) are powerful tools for exploring and learni...
Large decoder-only language models (LMs) can be largely improved in term...
Augmenting pretrained language models (LMs) with a vision encoder (e.g.,...
Dialogue systems can leverage large pre-trained language models and know...
As ultra-realistic face forgery techniques emerge, deepfake detection ha...
Large-scale vision-language pre-trained (VLP) models are prone to halluc...
Quantization is a technique to reduce the computation and memory cost of...
Natural language understanding (NLU) is the task of semantic decoding of...
The stance detection task aims to classify the stance toward given docum...
Cross-domain sentiment analysis aims to predict the sentiment of texts i...
Recent studies have shown great promise in applying graph neural network...
Self-supervised pre-training methods have brought remarkable breakthroug...
Existing knowledge-grounded dialogue systems typically use finetuned ver...
Deep learning (DL) models have achieved great success in many applicatio...
Code-switching is a speech phenomenon when a speaker switches language d...
Named entity recognition (NER) models generally perform poorly when larg...
We study Comparative Preference Classification (CPC) which aims at predi...
Multimodal abstractive summarization (MAS) models that summarize videos
...
Task-oriented compositional semantic parsing (TCSP) handles complex nest...
Information-seeking dialogue systems, including knowledge identification...
To diversify and enrich generated dialogue responses, knowledge-grounded...
The data scarcity in low-resource languages has become a bottleneck to
b...
Multilingual language models have shown decent performance in multilingu...
State-of-the-art abstractive summarization models generally rely on exte...
Existing works on multimodal affective computing tasks, such as emotion
...
Cross-domain named entity recognition (NER) models are able to cope with...
Many hardware vendors have introduced specialized deep neural networks (...
Despite the promising results of current cross-lingual models for spoken...
Task-oriented dialogue systems are either modularized with separate dial...
Despite the recent achievements made in the multi-modal emotion recognit...
Most emotion recognition methods tackle the emotion understanding task b...
Task-oriented dialogue systems use four connected modules, namely, Natur...
An increasing number of people in the world today speak a mixed-language...
Recently, fine-tuning pre-trained cross-lingual models (e.g., multilingu...
Nowadays, offensive content in social media has become a serious problem...
As an essential task in task-oriented dialog systems, slot filling requi...
Despite the great promise of Transformers in many sequence modeling task...
Personalized dialogue systems are an essential step toward better
human-...
Local dialects influence people to pronounce words of the same language
...
Existing models for cross-domain named entity recognition (NER) rely on
...
Most of the recent work in cross-lingual adaptation does not consider th...
Multiple convolutional neural network (CNN) classifiers have been propos...
Recently, data-driven task-oriented dialogue systems have achieved promi...
Despite the surging demands for multilingual task-oriented dialog system...
High performing deep neural networks come at the cost of computational
c...
In countries that speak multiple main languages, mixing up different
lan...
Time-lapse is a technology used to record the development of embryos dur...
This paper describes CAiRE's submission to the unsupervised machine
tran...
In this paper, we present an end-to-end empathetic conversation agent CA...