# Reward-Guided Decoding

CARD introduces a hierarchical framework for personalized text generation: it clusters users by shared stylistic patterns and trains group-specific LoRA adapters, derives lightweight user preference vectors through implicit preference learning that contrasts user-authored text with cluster-level generations, and injects personalization at inference only via low-rank logit corrections, ranking first in 10 of 12 settings across six LaMP and LongLaMP tasks and two metrics while remaining stable for low-resource users, across model scales, and in storage efficiency.

CARD introduces a hierarchical framework for personalized text generation: it clusters users by shared stylistic patterns and trains group-specific LoRA adapters, derives lightweight user preference vectors through implicit preference learning that contrasts user-authored text with cluster-level generations, and injects personalization at inference only via low-rank logit corrections, ranking first in 10 of 12 settings across six LaMP and LongLaMP tasks and two metrics while remaining stable for low-resource users, across model scales, and in storage efficiency.