RecRec: Latent Interests Recursive Reasoning for Sequential Recommendation
Rec is proposed, an RL-free framework that decouples reasoning from prediction, overcoming the fixed d-dimensional state bottleneck of prior methods, and points to a decoupled, multi-vector recipe that unleashes latent reasoning from the single-state bottleneck of prior methods.
Wenhao Deng, Junchen Fu, Han-Wen Du et al.
· arXiv.org · 0 citations