In this work, we introduce a model and beam-search training scheme, based on the work of Daume III and Marcu (2005), that extends seq2seq to learn global sequence scores.
Ranked #14 on Machine Translation on IWSLT2015 German-English
Directly reading documents and being able to answer questions from them is an unsolved challenge.
Ranked #6 on Question Answering on WikiQA
We propose a simple neural architecture for natural language inference.
Ranked #23 on Natural Language Inference on SNLI
We demonstrate that standard knowledge distillation applied to word-level prediction can be effective for NMT, and also introduce two novel sequence-level versions of knowledge distillation that further improve performance, and somewhat surprisingly, seem to eliminate the need for beam search (even when applied on the original teacher model).
Ranked #1 on Machine Translation on IWSLT2015 Thai-English
Such importance degree and text representation are calculated with multiple computational layers, each of which is a neural attention model over an external memory.
We present the Stanford Question Answering Dataset (SQuAD), a new reading comprehension dataset consisting of 100, 000+ questions posed by crowdworkers on a set of Wikipedia articles, where the answer to each question is a segment of text from the corresponding reading passage.
Approaches to multimodal pooling include element-wise product or sum, as well as concatenation of the visual and textual representations.
This paper explores the task of translating natural language queries into regular expressions which embody their meaning.