this post was submitted on 02 Jun 2024
21 points (100.0% liked)
Futurology
1807 readers
66 users here now
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
This seems interesting. For starters, I wonder if a model could be trained to spot logical errors in mathematical proofs. That in itself could be a useful tool for mathematicians, but it seems like an error detection model could also be used to improve existing text generation models through adversarial training.