Making maple syrup from sap is an illuminating experience. You have to reduce something like 40x the volume of sap to create the desired quantity of syrup.
I think the parent's point is that since price increases are capped, landlords cannot increase beyond a certain point in a single bumper year. To compensate, they may increase rents beyond what they would otherwise to be able to maintain a similar average increase. Citation is needed, of course.
The "bookcorpus" is from this paper by Zhu, Kiro, et. al. [0]. You can see the project web page here [1] which indicates you can crawl books from this site [2] to create your own. This repo seeks to replicate the original dataset [3]
There's also a huge variance in microphone quality for Android devices. The engineering focus to ensure uniform sound quality could take away from the team's focus in its early stages.
Has there been any good work applying causal inference explicitly to common NLP tasks like question-answering? Or to game-play? The lack of common-sense in models like GPT-3 would seem to indicate a huge opportunity, perhaps in tandem with COMET[0] or some other graph-based approach.
[0] https://arxiv.org/abs/1906.05317