Newest and Lowest Rated Papers
Advanced filtersTopic: AllExperimental
Topic
Topics are auto-detected from title, abstract, and metadata and may be imperfect.
Conceiving God
Unknown Author
Theology in the Flesh·2016
No abstract available
No ratings yet
View paper →Abolitionist Alternatives
Che Gossett
cabinetmagazine.org·2020
Black radicalism and the refusal of reform.
This text is part of “Imagination and the Carceral State,” a series of essays organized by Joshua Bennett for Cabinet’s Kiosk platform. For his introduction to the portfolio, see here. The other texts in the series include Bennett’s “Wherever We Are Gathered,” “No Man’s Land” by Elleza Kelley, and “Myth Lessons” by Matthew Spellberg.
No ratings yet
View paper →A Geometric Framework for Understanding Memorization in Generative Models
Brendan Leigh Ross, Hamidreza Kamkari, Tongzi Wu, Rasa Hosseinzadeh, Zhaoyan Liu, George Stein, Jesse C. Cresswell, Gabriel Loaiza-Ganem
arXiv·2024
As deep generative models have progressed, recent work has shown them to be capable of memorizing and reproducing training datapoints when deployed. These findings call into question the usability of generative models, especially in light of the legal and privacy risks brought about by memorization. To better understand this phenomenon, we propose the manifold memorization hypothesis (MMH), a geometric framework which leverages the manifold hypothesis into a clear language in which to reason about memorization. We propose to analyze memorization in terms of the relationship between the dimensionalities of (i) the ground truth data manifold and (ii) the manifold learned by the model. This framework provides a formal standard for "how memorized" a datapoint is and systematically categorizes memorized data into two types: memorization driven by overfitting and memorization driven by the underlying data distribution. By analyzing prior work in the context of the MMH, we explain and unify assorted observations in the literature. We empirically validate the MMH using synthetic data and image datasets up to the scale of Stable Diffusion, developing new tools for detecting and preventing generation of memorized samples in the process.
No ratings yet
View paper →On Forgetting and Stability of Score-based Generative models
Stanislas Strasman, Gabriel Cardoso, Sylvain Le Corff, Vincent Lemaire, Antonio Ocello
arXiv·2026
Understanding the stability and long-time behavior of generative models is a fundamental problem in modern machine learning. This paper provides quantitative bounds on the sampling error of score-based generative models by leveraging stability and forgetting properties of the Markov chain associated with the reverse-time dynamics. Under weak assumptions, we provide the two structural properties to ensure the propagation of initialization and discretization errors of the backward process: a Lyapunov drift condition and a Doeblin-type minorization condition. A practical consequence is quantitative stability of the sampling procedure, as the reverse diffusion dynamics induces a contraction mechanism along the sampling trajectory. Our results clarify the role of stochastic dynamics in score-based models and provide a principled framework for analyzing propagation of errors in such approaches.
No ratings yet
View paper →