As a grad student in the field, I 100% agree with the idea of research debt. Arxiv is great, but it has only exacerbated the problem. There's far too many paper

As a grad student in the field, I 100% agree with the idea of research debt. Arxiv is great, but it has only exacerbated the problem. There's far too many papers adding some small fluorish on an already complex architecture in order to push an accuracy score a couple of percentage points higher, and not that many with insightful examination of some more basic question, even if it's considered a detail.