TEST-TIME SELF-SUPERVISED GUIDANCE FOR DIFFUSION MODELS

Organization Name

qualcomm incorporated

Inventor(s)

Risheek Garrepalli of San Diego CA (US)

Yunxiao Shi of San Diego CA (US)

Hong Cai of San Diego CA (US)

Yinhao Zhu of La Jolla CA (US)

Shubhankar Mangesh Borse of San Diego CA (US)

Jisoo Jeong of San Diego CA (US)

Debasmit Das of San Diego CA (US)

Manish Kumar Singh of San Diego CA (US)

Rajeev Yasarla of San Diego CA (US)

Shizhong Steve Han of San Diego CA (US)

Fatih Murat Porikli of San Diego CA (US)

TEST-TIME SELF-SUPERVISED GUIDANCE FOR DIFFUSION MODELS

This abstract first appeared for US patent application 20240412493 titled 'TEST-TIME SELF-SUPERVISED GUIDANCE FOR DIFFUSION MODELS

Original Abstract Submitted

systems and techniques are provided for processing image data. according to some aspects, a computing device can generate a gradient (e.g., a classifier gradient using a trained classifier) associated with a current sample. the computing device can combine the gradient with an iterative model estimated score function or data associated with the current sample to generate a score function estimate. the computing device can predict, using the diffusion machine learning model and based on the score function estimate, a new sample.