paper-with-me

홈 › Papers

Self-Consuming Generative Models Go MAD

2023-07-04 · Sina AlEMohammad, Josue Casco-Rodriguez, Lorenzo Luzi, Ahmed Imtiaz Humayun, Hossein Babaei, Daniel LeJeune, Ali Siahkoohi, Richard G. Baraniuk

Seismic advances in generative AI algorithms for imagery, text, and other data types has led to the temptation to use synthetic data to train next-generation models. Repeating this process creates an autophagous (self-consuming) loop whose properties are poorly understood. We conduct a thorough analytical and empirical analysis using state-of-the-art generative image models of three families of autophagous loops that differ in how fixed or fresh real training data is available through the generations of training and in whether the samples from previous generation models have been biased to trade off data quality versus diversity. Our primary conclusion across all scenarios is that without enough fresh real data in each generation of an autophagous loop, future generative models are doomed to have their quality (precision) or diversity (recall) progressively decrease. We term this condition Model Autophagy Disorder (MAD), making analogy to mad cow disease.

📄 PDF Abstract BibTeX arXiv:2307.01850

Code (0)

등록된 구현이 없습니다.

Tasks

Diversity

Similar Papers 제목 키워드 기반

Self-Correcting Self-Consuming Loops for Generative Model Training

2024-02-11 · Nate Gillman, Michael Freeman, Daksh Aggarwal, Chia-Hong Hsu 외

As synthetic data becomes higher quality and proliferates on the internet, machine learning models are increasingly trained on a mix of human- and machine-generated data. Despite the successful stories of using synthetic…

Motion SynthesisRepresentation Learning

Convergence and Stability Analysis of Self-Consuming Generative Models with Heterogeneous Human Curation

2025-11-12 · Hongru Zhao, Jinwen Fu, Tuan Pham arxiv

Self-consuming generative models have received significant attention over the last few years. In this paper, we study a self-consuming generative model with heterogeneous preferences that is a generalization of the model…

Self-Consuming Generative Models with Adversarially Curated Data

2025-05-14 · Xiukun Wei, Xueru Zhang

Recent advances in generative models have made it increasingly difficult to distinguish real data from model-generated synthetic data. Using synthetic data for successive training of future model generations creates "sel…

Stabilizing Self-Consuming Diffusion Models with Latent Space Filtering

2025-11-16 · Zhongteng Cai, Yaxuan Wang, Yang Liu, Xueru Zhang arxiv

As synthetic data proliferates across the Internet, it is often reused to train successive generations of generative models. This creates a ``self-consuming loop" that can lead to training instability or \textit{model co…

Dual-branch PolSAR Image Classification Based on GraphMAE and Local Feature Extraction

2024-08-08 · Yuchen Wang, Ziyi Guo, Haixia Bi, Danfeng Hong 외

The annotation of polarimetric synthetic aperture radar (PolSAR) images is a labor-intensive and time-consuming process. Therefore, classifying PolSAR images with limited labels is a challenging task in remote sensing do…

Classificationimage-classificationImage ClassificationSelf-Supervised Learning