Congrats on the release! Finetuning a language model for decision classification (with probabilities) is already well-understood. What specifically changes in

Congrats on the release! Finetuning a language model for decision classification (with probabilities) is already well-understood. What specifically changes in the training objective with RLCD? Are its benefits isolated from Jev’s new architecture/parallelism?