Natural language is among the most accessible tools for explaining decis...
Modeling customer shopping intentions is a crucial task for e-commerce, ...
Answering complex questions often requires reasoning over knowledge grap...
We propose and study Complementary Concept Generation (CCGen): given a
c...
This paper presents a comprehensive and practical guide for practitioner...
Knowledge distillation has been shown to be a powerful model compression...
E-commerce query understanding is the process of inferring the shopping
...
E-commerce queries are often short and ambiguous. Consequently, query
un...
Graph neural network (GNN) pre-training methods have been proposed to en...
As training deep learning models on large dataset takes a lot of time an...
Recent research showed promising results on combining pretrained languag...
Predicting missing facts in a knowledge graph (KG) is crucial as modern ...
Recent research has shown the existence of significant redundancy in lar...
Self-training achieves enormous success in various semi-supervised and
w...
Adversarial regularization can improve model generalization in many natu...
Weak supervision has shown promising results in many natural language
pr...
The Lottery Ticket Hypothesis suggests that an over-parametrized network...
Adversarial training has been shown to improve the generalization perfor...
Existing curriculum learning approaches to Neural Machine Translation (N...
Reliable automatic evaluation of dialogue systems under an interactive
e...
Fine-tuned pre-trained language models can suffer from severe miscalibra...
Fine-tuned pre-trained language models (LMs) achieve enormous success in...
We study the open-domain named entity recognition (NER) problem under di...
Deep neural networks have been widely adopted in modern reinforcement
le...
Modern data acquisition routinely produce massive amounts of event seque...
Transfer learning has fundamentally changed the landscape of natural lan...
Many multi-domain neural machine translation (NMT) models achieve knowle...
Recently, with the help of deep learning models, significant advances ha...
This paper proposes a new meta-learning method -- named HARMLESS (HAwkes...
The learning rate warmup heuristic achieves remarkable success in stabil...
Deep neural networks have revolutionized many real world applications, d...
Optimal Transport (OT) naturally arises in many machine learning
applica...
Generative Adversarial Networks (GANs), though powerful, is hard to trai...
Adversarial training provides a principled approach for training robust
...