AI’s Memorization Crisis

O n Tuesday, researchers at Stanford and Yale revealed something that AI companies would prefer to keep hidden. Four popular ...
Tech Xplore

Discrete spatial diffusion models data while obeying scientific principles

Researchers at Los Alamos National Laboratory have developed a new approach that addresses the limitations of generative AI ...
Tech Xplore

One image is all robots need to find their way

While the capabilities of robots have improved significantly over the past decades, they are not always able to reliably and ...
GitHub

SVG: Latent Diffusion Model without Variational Autoencoder

SVG Autoencoder - Uses a frozen representation encoder with a residual branch to compensate the information loss and a learned convolutional decoder to transfer the SVG latent space to pixel space.
GitHub

[NeurIPS 2025 Spotlight] LaViDa:A Large Diffusion Language Model for Multimodal Understanding

Dec 16 2025 We released the preprint and Project Page for Sparse-LaViDa, an efficient optimization technique for training and sampling from unified multi-modal dLLMs based on LaViDa. Oct 2025: We ...
IEEE

Quantitative Phase Imaging Denoising Based on Denoising Diffusion Probabilistic Models

Abstract: Quantitative Phase Imaging (QPI) has been shown to complement established fluorescence microscopy as well as objective measurements of morphology and dynamics for cellular tissue studies.
IEEE

Denoising Diffusion Probabilistic Steganography Based on Standardized Secret Noise

Abstract: Generative steganography based on diffusion model is a technique that directly uses secret information to generate a stego image by diffusion model. Due to its strong resistance to ...