Issue 2026-11-15 · Industry · 研究
Researchers question effectiveness of double-blind review
Reddit discussion argues double-blind review in ML conferences often works like single-blind, with reviewers able to see author names and affiliations. The poster claims to have seen student reviewers reject papers to improve their own chances, and questions whether well-connected labs get fair evaluation. No quantitative evidence was provided.
Read original ↗