arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.19157cs.CL

通过对比稀疏自编码器的特质激活路由实现面向级人格控制

Facet-Level Persona Control by Trait-Activated Routing with Contrastive SAE for Role-Playing LLMs

Nagoya University（名古屋大学）；Kyoto University（京都大学）

Wenqiu Tang, Zhen Wan, Takahiro Komamizu, Ichiro Ide

备注：
Accepted in PAKDD 2026 special session on Data Science :Foundation and Applications
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出一种基于对比稀疏自编码器的特质激活路由框架，用于角色扮演大语言模型的人格控制，通过构建平衡监督的语料库，实现稳定的人格表现和输出质量。

摘要：

角色扮演代理（RPAs）中的人格控制通常通过免训练方法实现，如通过提示或检索增强生成注入人格描述和记忆，或通过在特定人格语料库上进行监督微调（SFT）。虽然SFT有效，但需要标记的人格数据和重新训练新角色，限制了灵活性。相比之下，提示和RAG基于的信号易于应用，但在长对话中可能被稀释，导致漂移和不一致的人格行为。为此，我们提出了一种对比稀疏自编码器（SAE）框架，学习与大五人格30个特质模型对齐的面向级人格控制向量。构建了一个新的15,000样本的泄漏控制语料库，为每个特质提供平衡的监督。学习的向量被整合到模型的残差空间中，并通过特质激活路由模块动态选择，实现精确且可解释的人格引导。在大型语言模型（LLMs）上的实验表明，所提方法在上下文化设置中保持稳定的人格保真度和输出质量，优于对比激活添加（CAA）和仅提示基线。结合SAE+提示配置实现了最佳整体性能，证实了对比训练的潜在向量可以增强人格控制同时保持对话连贯性。数据集可在：https://github.com/lunat5078/BigFive-Personality-Facets-Dataset获取。

英文摘要：

Personality control in Role-Playing Agents (RPAs) is commonly achieved via training-free methods that inject persona descriptions and memory through prompts or retrieval-augmented generation, or via supervised fine-tuning (SFT) on persona-specific corpora. While SFT can be effective, it requires persona-labeled data and retraining for new roles, limiting flexibility. In contrast, prompt- and RAG-based signals are easy to apply but can be diluted in long dialogues, leading to drifting and sometimes inconsistent persona behavior. To address this, we propose a contrastive Sparse AutoEncoder (SAE) framework that learns facet-level personality control vectors aligned with the Big Five 30-facet model. A new 15,000-sample leakage-controlled corpus is constructed to provide balanced supervision for each facet. The learned vectors are integrated into the model's residual space and dynamically selected by a trait-activated routing module, enabling precise and interpretable personality steering. Experiments on Large Language Models (LLMs) show that the proposed method maintains stable character fidelity and output quality across contextualized settings, outperforming Contrastive Activation Addition (CAA) and prompt-only baselines. The combined SAE+Prompt configuration achieves the best overall performance, confirming that contrastively trained latent vectors can enhance persona control while preserving dialogue coherence. Dataset is available at: https://github.com/lunat5078/BigFive-Personality-Facets-Dataset

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑