Research and ACM SRC Posters Archive

DiffPro: Joint Timestep and Layer-Wise Precision Optimization for Efficient Diffusion Inference

Poster Type: Research Posters

Author: Farhana Amin (Virginia Tech), Kanchon Gharami (Embry-Riddle Aeronautical University), Dimitrios S. Nikolopoulos (Virginia Tech)

Supervisor:

Abstract: DiffPro is a simple framework to speed up and shrink diffusion models while preserving image quality. It combines layer-wise quantization, guided by a manifold-based sensitivity check with adaptive timestep selection. Compared with quantization-only or sampling-only baselines, this joint strategy yields better FID–memory trade-offs. We evaluate on MNIST, CIFAR-10, and CelebA using both PTQ and QAT, showing meaningful size reductions (e.g., ~34 MB on a CIFAR-10 setup) with competitive FID. Current limits include occasional mis-ranking of a few blocks at high noise and a focus on unconditioned models. Next, we will extend DiffPro to text/class-conditioned diffusion, replace hand-tuned thresholds with a budgeted optimizer that co-selects per-layer bit-widths and timesteps using per-timestep sensitivity, and incorporate hardware-aware costs.

Best Poster Finalist (BP): no
Poster: PDF
Poster Summary: PDF

Back to Poster Archive Listing