Behaviorally Grounded User Profiles from the Wild for Personalized Alignment and Multi-Perspective Reasoning

基于真实行为的开放式用户画像：用于个性化对齐与多视角推理

Abstract: Persona-driven techniques increasingly adapt large language models (LLMs) to diverse contexts. However, existing methods predominantly rely on rigid, synthetic personas that flatten individual variation, rely on stereotypes, and miss the nuanced signals driving actual human preferences.

摘要： 基于角色驱动的技术正日益将大语言模型（LLMs）适配到各种不同的场景中。然而，现有的方法主要依赖于僵化的合成角色，这些角色往往抹杀了个人差异，依赖刻板印象，并忽略了驱动人类真实偏好的细微信号。

We introduce profile behavioral grounding, a framework for extracting open-ended, high-fidelity user profiles directly from authentic, anonymized social media posts. We evaluate these profiles across two paradigms: train-time personalization via supervised finetuning (SFT) and non-parametric test-time multi-perspective reasoning.

我们引入了“用户画像行为基础”（profile behavioral grounding）框架，该框架旨在直接从真实的、匿名化的社交媒体帖子中提取开放式、高保真的用户画像。我们在两种范式下对这些画像进行了评估：通过监督微调（SFT）进行的训练时个性化，以及非参数化的测试时多视角推理。

Across complex recommendation and open-ended query benchmarks, behaviorally grounded profiles consistently improve base models and outperform synthetic profile baselines, driving stronger parametric alignment and enabling richer, multifaceted reasoning.

在复杂的推荐系统和开放式查询基准测试中，基于行为的用户画像始终能提升基础模型的效果，并优于合成画像基准，从而推动了更强的参数对齐，并实现了更丰富、更多维度的推理。

Our findings establish open-ended, behavior-derived profiles as a highly diverse and effective foundation for the next generation of personalized language systems. Our code base is available at this https URL.

我们的研究结果表明，这种基于行为的开放式用户画像，是下一代个性化语言系统高度多样化且有效的基石。我们的代码库已在链接中提供。