Author name: Etienne Gauthier, Francis Bach, Michael I. Jordan

arXiv:2605.04266v1 Announce Type: cross
Abstract: Reinforcement learning from human feedback (RLHF) typically assumes a static or non-strategic reward model (RM). In iterative deployment, however, the policy generates the data on which the RM is retra…

arXiv:2510.04318v2 Announce Type: replace-cross
Abstract: Traditional conformal prediction methods construct prediction sets such that the true label falls within the set with a user-specified coverage level. However, poorly chosen coverage levels can…