← 研踪 PaperTrailarXiv:2607.21067 · v2
PrefReward: Learning User Preference Matrix for Personalized Text Generation
首次提交 2026-07-23 17:00
版本更新 2026-11-10 15:43
雷达观测 2026-11-11 13:08
摘要
Large Language Models (LLMs) have demonstrated remarkable ability in generating personalized content by leveraging user histories and contextual cues. However, most existing personalization approaches rely on implicit representations within model parameters, making it difficult to interpret user-specific preferences or effectively handle long-context dependencies. To address these challenges, we propose PrefReward, a novel preference-aware generative framework that explicitly models user styles through a structured preference matrix and integrates it into the decoding process as a reward signal. PrefReward consists of two stages: (1) extracting a user-specific preference matrix that summarizes individual stylistic tendencies, and (2) using the matrix to guide generation via a KL-divergence-based reward function. Experiments on the LongLaMP dataset show that PrefReward outperforms retrieval-based baselines and context steering method in both generation quality and personalization interpretability.
作者备注：added several analyses and visualizations, typos corrected
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-10 15:43 · 最新观测版本
PrefReward: Learning User Preference Matrix for Personalized Text Generation
Large Language Models (LLMs) have demonstrated remarkable ability in generating personalized content by leveraging user histories and contextual cues. However, most existing personalization approaches rely on implicit representations within model parameters, making it difficult to interpret user-specific preferences or effectively handle long-context dependencies. To address these challenges, we propose PrefReward, a novel preference-aware generative framework that explicitly models user styles through a structured preference matrix and integrates it into the decoding process as a reward signal. PrefReward consists of two stages: (1) extracting a user-specific preference matrix that summarizes individual stylistic tendencies, and (2) using the matrix to guide generation via a KL-divergence-based reward function. Experiments on the LongLaMP dataset show that PrefReward outperforms retrieval-based baselines and context steering method in both generation quality and personalization interpretability.
查看此版本解读 →查看此版本原文 ↗