When Preferences Diverge: Aligning Diffusion Models with Minority-Aware Adaptive DPO

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Lingfan, Liu, Chen, Xu, Chengming, Hu, Kai, Luo, Donghao, Wang, Chengjie, Fu, Yanwei, Yao, Yuan
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916658992381952
author Zhang, Lingfan
Liu, Chen
Xu, Chengming
Hu, Kai
Luo, Donghao
Wang, Chengjie
Fu, Yanwei
Yao, Yuan
author_facet Zhang, Lingfan
Liu, Chen
Xu, Chengming
Hu, Kai
Luo, Donghao
Wang, Chengjie
Fu, Yanwei
Yao, Yuan
contents In recent years, the field of image generation has witnessed significant advancements, particularly in fine-tuning methods that align models with universal human preferences. This paper explores the critical role of preference data in the training process of diffusion models, particularly in the context of Diffusion-DPO and its subsequent adaptations. We investigate the complexities surrounding universal human preferences in image generation, highlighting the subjective nature of these preferences and the challenges posed by minority samples in preference datasets. Through pilot experiments, we demonstrate the existence of minority samples and their detrimental effects on model performance. We propose Adaptive-DPO -- a novel approach that incorporates a minority-instance-aware metric into the DPO objective. This metric, which includes intra-annotator confidence and inter-annotator stability, distinguishes between majority and minority samples. We introduce an Adaptive-DPO loss function which improves the DPO loss in two ways: enhancing the model's learning of majority labels while mitigating the negative impact of minority samples. Our experiments demonstrate that this method effectively handles both synthetic minority data and real-world preference data, paving the way for more effective training methodologies in image generation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2503_16921
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle When Preferences Diverge: Aligning Diffusion Models with Minority-Aware Adaptive DPO
Zhang, Lingfan
Liu, Chen
Xu, Chengming
Hu, Kai
Luo, Donghao
Wang, Chengjie
Fu, Yanwei
Yao, Yuan
Computer Vision and Pattern Recognition
Artificial Intelligence
In recent years, the field of image generation has witnessed significant advancements, particularly in fine-tuning methods that align models with universal human preferences. This paper explores the critical role of preference data in the training process of diffusion models, particularly in the context of Diffusion-DPO and its subsequent adaptations. We investigate the complexities surrounding universal human preferences in image generation, highlighting the subjective nature of these preferences and the challenges posed by minority samples in preference datasets. Through pilot experiments, we demonstrate the existence of minority samples and their detrimental effects on model performance. We propose Adaptive-DPO -- a novel approach that incorporates a minority-instance-aware metric into the DPO objective. This metric, which includes intra-annotator confidence and inter-annotator stability, distinguishes between majority and minority samples. We introduce an Adaptive-DPO loss function which improves the DPO loss in two ways: enhancing the model's learning of majority labels while mitigating the negative impact of minority samples. Our experiments demonstrate that this method effectively handles both synthetic minority data and real-world preference data, paving the way for more effective training methodologies in image generation tasks.
title When Preferences Diverge: Aligning Diffusion Models with Minority-Aware Adaptive DPO
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2503.16921