Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now


Alibaba’s deep learning model beats human in latest test

The robots are taking over...

Alibaba’s DAMO Academy Language Technology Lab which focuses on Natural Language Processing (NLP), has developed a deep-learning model that scored higher than a human being on a Q&A task in Microsoft’s latest machine reading comprehension test.

The MS MARCO test requires the challengers to go through about 1,000 documents based on Microsoft’s Bing search dataset to retrieve relevant passages first, and then discover or even self-develop the right answers to address the designated questions.

According to the company, the achievement reveals machines’ potential in understanding large volumes of complex materials as well as answering complicated questions in real life. It says the technology could be applied to everything from intelligent chatbots to search engines that can come back with direct answers rather than links of webpages.

Alibaba is now looking at adding NLP to its Cloud computing platform enabling businesses in retail, tourism and public services that involve Q&A tasks to benefit from the technology.

“We are also exploring a multi-module model that combines features of NLP, speech AI and machine translation,” said Dr. Luo Si, leader of the Alibaba DAMO Academy Language Technology Lab. “With such a model, users with different languages can communicate more freely with each other and interact with the machine in real-time without language barriers.”