Web search is fundamentally multimodal and multihop. Often, even before
...
Pre-training text representations has recently been shown to significant...
In this paper, we introduce XGLUE, a new benchmark dataset to train
larg...
We present MT-DNN, an open-source natural language understanding (NLU)
t...
The web contains a vast corpus of HTML tables. They can be used to provi...
Tables extracted from web documents can be used to directly answer many ...
Commonsense question answering aims to answer questions which require
ba...
We present a generative model to map natural language questions into SQL...