paper-with-me

Papers

Latent Diffusion U-Net Representations Contain Positional Embeddings and Anomalies

2025-04-09 · Jonas Loos, Lorenz Linhardt

Diffusion models have demonstrated remarkable capabilities in synthesizing realistic images, spurring interest in using their representations for various downstream tasks. To better understand the robustness of these representations, we analyze popular Stable Diffusion models using representational similarity and norms. Our findings reveal three phenomena: (1) the presence of a learned positional embedding in intermediate representations, (2) high-similarity corner artifacts, and (3) anomalous high-norm artifacts. These findings underscore the need to further investigate the properties of diffusion model representations before considering them for downstream tasks that require robust features. Project page: https://jonasloos.github.io/sd-representation-anomalies

📄 PDF Abstract BibTeX arXiv:2504.07008

Code (1)

JonasLoos/sd-representation-anomalies 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

GLASS: Guided Latent Slot Diffusion for Object-Centric Learning

2025-01-01 · CVPR 2025 1 · Krishnakant Singh, Simone Schaub-Meyer, Stefan Roth

Object-centric learning aims to decompose an input image into a set of meaningful object files (slots). These latent object representations enable a variety of downstream tasks. Yet, object-centric learning struggles…

Conditional Image GenerationImage GenerationObjectObject Discovery

Compositional Discrete Latent Code for High Fidelity, Productive Diffusion Models

2025-07-16 · Samuel Lavoie, Michael Noukhovitch, Aaron Courville arxiv

We argue that diffusion models' success in modeling complex distributions is, for the most part, coming from their input conditioning. This paper investigates the representation used to condition diffusion models from th…

Self-Supervised LearningText-to-Image Generation

DDMI: Domain-Agnostic Latent Diffusion Models for Synthesizing High-Quality Implicit Neural Representations

2024-01-23 · Dogyun Park, Sihyeon Kim, Sojin Lee, Hyunwoo J. Kim

Recent studies have introduced a new class of generative models for synthesizing implicit neural representations (INRs) that capture arbitrary continuous signals in various domains. These models opened the door for domai…

3D Shape GenerationImage GenerationVideo Generation

EnergyMoGen: Compositional Human Motion Generation with Energy-Based Diffusion Model in Latent Space

2024-12-19 · CVPR 2025 1 · Jianrong Zhang, Hehe Fan, Yi Yang

Diffusion models, particularly latent diffusion models, have demonstrated remarkable success in text-driven human motion generation. However, it remains challenging for latent diffusion models to effectively compose mult…

Motion GenerationSemantic Composition

Not Only Text: Exploring Compositionality of Visual Representations in Vision-Language Models

2025-03-21 · CVPR 2025 1 · Davide Berasi, Matteo Farina, Massimiliano Mancini, Elisa Ricci 외

Vision-Language Models (VLMs) learn a shared feature space for text and images, enabling the comparison of inputs of different modalities. While prior works demonstrated that VLMs organize natural language representation…