NeurIPS 2020

Learning to Prove Theorems by Learning to Generate Theorems

Meta Review

We have a very thorough review for this paper giving it a rating of 6. The topic seems relevant and important but the scores are borderline. The main issue seems to be properly placing the work in the context of previous work with proper comparisons. There were also some complaints about the clarity of the presentation. The paper would greatly benefit from a rewrite and probably extended experiments taking these reviews into consideration.