Researchers seek AI pre-submission reviews for CVPR papers
AI summary
A Reddit user asked other CVPR researchers to share pre-submission reviews generated by Claude or ChatGPT. The poster said those reviews repeatedly called the work “Borderline” with limited novelty, even after revisions to the issues identified by the AI.
In-depth insight
The discussion provides a first-hand community signal that researchers may use Claude or ChatGPT for pre-submission feedback on academic papers, although it does not establish how common the practice is.
So what was your pre-submission review from Claude or ChatGPT?
For me it’s always "Boarderline" with limited novelty. No matter how many time I fix the issues reported by AI.