ICML Author Removal Policy Traps PhD Student: Keep List or Withdraw Paper

A PhD student on Reddit is sharing a high-stakes dilemma at the International Conference on Machine Learning (ICML). After the abstract deadline, several industry coauthors asked to be removed because they missed their employer’s internal approval window for publishing. The student hadn't explicitly asked permission before adding them. The student proactively contacted the program committee (PC) chairs in January, obtained written consent from all coauthors, and received explicit approval from the chairs to remove the names. The chairs confirmed they would implement the change.

Four months later, the paper was accepted—with the original author list still in place. At camera-ready follow-up, the PC chairs reversed their decision, citing a blanket policy: no author list changes allowed unless the paper is withdrawn. The student faces an impossible choice: keep coauthors who shouldn't be listed (potentially violating ethical guidelines) or withdraw the accepted paper, missing the Neurips deadline. The incident highlights the tension between bureaucratic inflexibility and real-world authoring complexities in top-tier conferences.