[27] Danqi Chen - Neural Reading Comprehension and Beyond
The Thesis Review
The Interaction Between Memory Networks and Neural Reading Comprehension Models
Dr. Darkheur has been working on memory networks for like one year or two years at that point. Memory networks are mostly like using like backwards encoding. So it actually doesn't really work that well only for like let's say, called type of questions. And also they try to use very similar heuristics to pick out like the memories. For example, top key memories that have like more overlap with the questions,. This kind of a heuristic. You can think of about the retrieval components of Dr.Darkheur as just a very natural step of that kind of memory retrieval. And also with much more sophisticated models for the reader part.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
