Research
Advancing how machines understand and generate human language across speech, text, and translation.
TextWorld is a text-based game generator and sandbox learning environment for reinforcement learning agents. Watch this video to get started with using TextWorld! Learn more about the competition at aka.ms/textworld-challenge
Language embedding is a process of mapping symbolic natural language text (for example, words, phrases and sentences) to semantic vector representations. This is fundamental to deep learning approaches to natural language understanding (NLU). It is…
We introduce a new corpus of descriptions of Xbox avatars created by actual gamers. Each avatar is specified by 19 attributes, including clothing and body type, allowing for more than 10^20 possibilities. Using Amazon Mechanical…
Building a computer system to automatically solve math word problems written in natural language. SigmaDolphin is a project initiated in early 2013 at Microsoft Research Asia, with the primary goal of building a computer intelligent…
As part of the Eighth Dialog System Technology Challenge (DSTC8), Microsoft Research and Tsinghua University are hosting a track intended to foster progress in two important aspects of dialog systems: dialog complexity and scaling to…
Episode 64, February 20, 2019 – Dr. Layla El Asri talks about the particular challenges she and other scientists face in building sophisticated dialogue systems that lay the foundation for talking machines. She also explains…
