Persona-Aware Alignment Framework for Personalized Dialogue Generation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Li, Guanrong, Liu, Xinyu, Wu, Zhen, Dai, Xinyu
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909900596051968
author Li, Guanrong
Liu, Xinyu
Wu, Zhen
Dai, Xinyu
author_facet Li, Guanrong
Liu, Xinyu
Wu, Zhen
Dai, Xinyu
contents Personalized dialogue generation aims to leverage persona profiles and dialogue history to generate persona-relevant and consistent responses. Mainstream models typically rely on token-level language model training with persona dialogue data, such as Next Token Prediction, to implicitly achieve personalization, making these methods tend to neglect the given personas and generate generic responses. To address this issue, we propose a novel Persona-Aware Alignment Framework (PAL), which directly treats persona alignment as the training objective of dialogue generation. Specifically, PAL employs a two-stage training method including Persona-aware Learning and Persona Alignment, equipped with an easy-to-use inference strategy Select then Generate, to improve persona sensitivity and generate more persona-relevant responses at the semantics level. Through extensive experiments, we demonstrate that our framework outperforms many state-of-the-art personalized dialogue methods and large language models.
format Preprint
id arxiv_https___arxiv_org_abs_2511_10215
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Persona-Aware Alignment Framework for Personalized Dialogue Generation
Li, Guanrong
Liu, Xinyu
Wu, Zhen
Dai, Xinyu
Computation and Language
Artificial Intelligence
Personalized dialogue generation aims to leverage persona profiles and dialogue history to generate persona-relevant and consistent responses. Mainstream models typically rely on token-level language model training with persona dialogue data, such as Next Token Prediction, to implicitly achieve personalization, making these methods tend to neglect the given personas and generate generic responses. To address this issue, we propose a novel Persona-Aware Alignment Framework (PAL), which directly treats persona alignment as the training objective of dialogue generation. Specifically, PAL employs a two-stage training method including Persona-aware Learning and Persona Alignment, equipped with an easy-to-use inference strategy Select then Generate, to improve persona sensitivity and generate more persona-relevant responses at the semantics level. Through extensive experiments, we demonstrate that our framework outperforms many state-of-the-art personalized dialogue methods and large language models.
title Persona-Aware Alignment Framework for Personalized Dialogue Generation
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2511.10215