arXiv Computer Vision By Wenshuai Yao, Wenyong Zhou

When Guidance Goes Off-Scale: Recalibrating Diffusion Transformers under Analog Compute-in-Memory Nonidealities

Read the original on arXiv Computer Vision →

arXiv:2608. 19644v1 Announce Type: new Abstract: Diffusion Transformers (DiTs) incur high memory traffic and energy costs because sampling repeatedly evaluates large denoisers dominated by linear operations.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.