Xiao Yang

TL;DR: The results show that learning from data automatically labeled by user engagement signals achieves significant accuracy gains in a production deep learning system, when measured on both the sequence labeling task as well as on user facing results produced by the system end-to-end.

TL;DR: In this paper, the authors present a process for operating an intelligent automated assistant that determines whether the second parsing result corresponds to a data item in a knowledge base, and in accordance with determining that the second parser corresponds to the data item, the process ranks the second parse result higher than the first parsing result.

TL;DR: In this paper, the authors propose a novel architecture that jointly solves the NER and EL tasks by combining them in a joint reranking module, and show that their proposed framework improves NER accuracy by up to 3.13% and EL accuracy by 3.6% in F1 score.

TL;DR: In this paper, the authors describe a system and a process for processing natural language input, which is based on a named entity model and the natural language inputs, where the second value defines a parameter for a task corresponding to the input.

TL;DR: An architecture with novel features that jointly solves the recognition of named entities and the resolution to their canonical forms and it is shown that by combining NER and EL information in a joint reranking module, the proposed framework improves accuracy in both tasks.