RL post-training makes generative recommenders better rankers but worse retrievers: Office hit@500 falls from 52.84% to 47.69%
Generative recommenders are now commonly post-trained with reinforcement learning after supervised fine-tuning, but gains are usually reported on the first ten results. Exact full-catalogue ranking on five datasets shows that RL post-training raises top-of-list accuracy while lowering recall deeper in the candidate list: on Office, NDCG@10 rises from 11.59% to 12.55% while hit@500 falls from 52.84% to 47.69%. The cause is that RL pushes down every sampled candidate except the logged target, though those candidates are often what similar users choose next; one anchoring term outside the rollout restores recall to the supervised level while keeping the top-of-list gain.