Chunyang Xiao

We propose an approach for semantic parsing that uses a recurrent neural network to map a natural language question into a logical form representation of a KB query. Expand

We introduce \emph{LL-RNNs} (Log-Linear RNNs), an extension of Recurrent Neural Networks that replaces the softmax output layer by a log-linear output layer, of which softmax is a special case. Expand

We propose to exploit various sources of prior knowledge: the well-formedness of the logical forms is modeled by a weighted context-free grammar; the likelihood that certain entities present in the input utterance are also present in a logical form is modeling by weighted finite-state automata. Expand

We develop a generic approach for restricting the predictions of a seq2seq model to grammatically permissible continuations, which leads to a 74% speed-up on an in-house dataset with a large vocabulary, compared to a neural model without grammatical restrictions. Expand

In an online decision problem, one makes decisions often with a pool of decision sequence called experts but without knowledge of the future. Expand

Learning embeddings of words and knowledge base elements is a promising approach for open domain question answering. Expand

Our goal in this thesis is to build a system that answers a natural language question (NL) by representing its semantics as a logical form (LF) and then computing the answer by executing the LF over a knowledge base. Expand