CS Events

By Year By Month By Week Today Jump to month Search

January February March April May June July August September October November December 2024 2025 2026 Jump to month

Qualifying Exam Antithetic Noise in Diffusion Models
Export Event Save iCal Send to Google Calendar Send to Yahoo Calendar Send to Outlook Live Send to Microsoft Outlook Save iCal Preserve formatting in description (only supported in some calendar applications) Close Friday, November 07, 2025, 05:00pm - 06:30pm

Speaker: Jing Jia

Location : CoRE 305

Committee: 

Professor Peng Zhang

Professor Ruixiang Tang

Professor Hao Wang

Professor Yipeng Huang

Event Type: Qualifying Exam

Abstract: We initiate a systematic study of antithetic initial noise in diffusion models. Across unconditional models trained on diverse datasets, text-conditioned latent diffusion models, and diffusion posterior samplers, we find that pairing each initial noise with its negation consistently yields strongly negatively correlated samples. To explain this phenomenon, we combine experiments and theoretical analysis, leading to a symmetry conjecture that the learned score function is approximately affine antisymmetric (odd symmetry up to a constant shift), and provide evidence supporting it. Leveraging this negative correlation, we enable two applications: (1) enhancing image diversity in models like Stable Diffusion without quality loss, and (2) sharpening uncertainty quantification (e.g., up to 90% narrower confidence intervals) when estimating downstream statistics. Building on these gains, we extend the two-point pairing to a randomized quasi-Monte Carlo estimator, which further improves estimation accuracy. Our framework is training-free, model-agnostic, and adds no runtime overhead.Link: https://arxiv.org/pdf/2506.06185

: 

Contact  Professor Peng Zhang

Zoom Link: https://rutgers.zoom.us/j/94679932398?pwd=L2oQPYbpb3Qc9ZkazRweQiQrhSHnCR.1