[27] Danqi Chen - Neural Reading Comprehension and Beyond
The Thesis Review
The Benefits of Generic Datasets for Reasoning
I definitely think there has been a lot of exciting progress in the last few years. And still, there are still challenging problems to solve as well. I'm actually still working on the projects related to these aspects. If we usually want to build one single unified model that can actually even generalize to out of domain distributions, I think it's still like an open problem. Actually, I do like the idea of like creating datasets. For example, like a market hub reasoning or like a preference reasoning or a new macro is kind of datasets.
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.
