paper-with-me

Papers

Balancing Fidelity and Diversity in Diffusion Models via Symmetric Attention Decomposition: Hopfield Perspective

2026-05-26 · Hyunmin Cho, Woo Kyoung Han, Kyong Hwan Jin arxiv

We characterize the pre-softmax attention matrix $\mathbf{QK^\top}$ in transformers as an associative memory matrix encoding pairwise associations between input features. By decomposing this matrix into its symmetric and skew-symmetric parts, we interpret the symmetric component as governing the structure of the energy landscape, and the skew-symmetric component as driving circulation on that landscape. Leveraging the energy formulation induced by the symmetric component, we derive Hopfield-style stability measures that quantify the stability of retrieved features. We observe meaningful correlations between Hopfield-style stability measures and the fidelity-diversity trade-offs in generation. Finally, we propose a controllable knob to modulate this trade-off by modifying the circulation of the underlying dynamics. Code is available at our GitHub (https://github.com/hyeon-cho/Attention-Symmetric-Decomposition).

📄 PDF Abstract BibTeX arXiv:2605.27476

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Boundary-Constrained Diffusion Models for Floorplan Generation: Balancing Realism and Diversity

2026-02-02 · Leonardo Stoppani, Davide Bacciu, Shahab Mokarizadeh arxiv

Diffusion models have become widely popular for automated floorplan generation, producing highly realistic layouts conditioned on user-defined constraints. However, optimizing for perceptual metrics such as the Fréchet I…

BokehDiff: Neural Lens Blur with One-Step Diffusion

2025-07-24 · Chengxuan Zhu, Qingnan Fan, Qi Zhang, Jinwei Chen 외 arxiv

We introduce BokehDiff, a novel lens blur rendering method that achieves physically accurate and visually appealing outcomes, with the help of generative diffusion prior. Previous methods are bounded by the accuracy of d…

Depth Estimation

Class-Balancing Diffusion Models

2023-04-30 · CVPR 2023 1 · Yiming Qin, Huangjie Zheng, Jiangchao Yao, Mingyuan Zhou 외

Diffusion-based models have shown the merits of generating high-quality visual data while preserving better diversity in recent studies. However, such observation is only justified with curated data distribution, where t…

Diversity

AI for Cultural Heritage Textiles: Fine-Tuned Latent Diffusion for Novel Ulos Motif Synthesis

2026-07-06 · Humasak Tommy Argo Simanjuntak, Jesika Purba, Sitogab Girsang, Widya Manurung 외 arxiv

Preserving and revitalising traditional textiles such as Ulos, a cultural heritage of the Batak ethnic group in North Sumatra, Indonesia, requires balancing fidelity to tradition with innovative approaches that meet cont…

DreamBlend: Advancing Personalized Fine-tuning of Text-to-Image Diffusion Models

2024-11-28 · Shwetha Ram, Tal Neiman, Qianli Feng, Andrew Stuart 외

Given a small number of images of a subject, personalized image generation techniques can fine-tune large pre-trained text-to-image diffusion models to generate images of the subject in novel contexts, conditioned on tex…

DiversityImage GenerationPersonalized Image Generation