NeurIPS Peer Review Controversy: Perfect Rebuttals Become a Reason for Rejection

CSProfKGD · x · 2026-08-07

A rant about the NeurIPS 2026 paper review process is resonating within the academic community. The original poster shared their roller-coaster review experience: their paper's initial ratings were 543, which later moved to 554.

The Area Chair (AC) initially gave a lean reject rating and listed 9 specific concerns. During the rebuttal phase, the author's team systematically addressed all 9 issues. However, the AC's subsequent feedback was absurd: while acknowledging that the theoretical and novelty concerns were resolved, the AC pivoted to criticize the authors for "changing too much," maintaining the lean reject decision.

The poster expressed frustration, noting that under the current peer review system, it seems rebuttals simply "cannot work too well."

Original post →

More from Fun

Fun channel →