ServiceNow Research

Diffusion models

Unifying Autoregressive and Diffusion-Based Sequence Generation
We present significant extensions to diffusion-based sequence generation models, blurring the line with autoregressive language models. …
Adaptive Diffusion Denoised Smoothing : Certified Robustness via Randomized Smoothing with Differentially Private Guided Denoising Diffusion
We propose Adaptive Diffusion Denoised Smoothing, a method for certifying the predictions of a vision model against adversarial …
Unifying Autoregressive and Diffusion-Based Sequence Generation
We take significant steps toward unifying autoregressive and diffusion-based sequence generation by extending the SEDD discrete …
Exploring validation metrics for offline model-based optimisation with diffusion models
In model-based optimisation (MBO) we are interested in using machine learning to design candidates that maximise some measure of reward …
Are Diffusion Models Vision-And-Language Reasoners?
Text-conditioned image generation models have recently shown immense qualitative success using denoising diffusion processes. However, …
Exploring the Design Space of Generative Diffusion Processes for Sparse Graphs
We extend score-based generative diffusion processes (GDPs) to sparse graphs and other inherently discrete data, with a focus on …
Masked Conditional Video Diffusion for Prediction, Generation, and Interpolation
Video prediction is a challenging task. The quality of video frames from current state-of-the-art (SOTA) generative models tends to be …