AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Importance of Question Answering and Dialogue in Dialogue Systems
So I basically use a co-cur, so like a natural next step after we solve this like a single passage, single term reading comprehension problem. Can we actually build models to actually also just kind of context dependent questions? So this is basically the main idea of the co-cur project or co-cur dataset. And it's actually much easier to evaluate compared to like, let's say, chit chat style dialogue systems,. But still a very important, could it be a single sort of important component in like dialogue systems?