Dealing with Followup Questions in Alexa

How can we deal with common expectations of conversation, like followup questions?  See the reference to the upcoming paper.  The blog linked to often has interesting, but technical results.  Still, often useful to scan, since these are aimed at skill and architecture developers, to understand the difficulty involved.  Also, how much context is meant to be in memory for followup?

Amazon researchers share how AI helps Alexa answer follow-up questions   By  Kyle Wiggers  in VentureBeat

Humans have an easy time with context carryover — the ability to track references through rounds of conversation (like inferring the meaning of “there” in a follow-up to “How is the weather in San Francisco?”) — but it’s beyond the conversational repertoire of most smart speakers and virtual assistants. That’s nothing a little artificial intelligence (AI) can’t fix, though, and in a blog post today researchers at Amazon wrote about the progress they’ve made with Alexa.

During a typical chat with Alexa, users invoke multiple apps — skills, in Amazon’s parlance — in successive questions. Currently, Alexa analyzes the content of each query according to its domain (type of skill), intent (function of the skill), and slot (a variable the skill acts upon). But skills often repurpose slots like “town” and “city,” which poses an obvious problem — if a user asks for directions and follows up with a question about a restaurant’s location, how’s Alexa to know which thread to reference in its answer?

Gupta and colleagues describe a solution in the paper “Contextual Slot Carryover for Disparate Schemas,” a neural network that automatically learns to map one skill’s slots to another’s. The findings will be presented at the upcoming Interspeech conference in Hyderabad, India in September. .... "