Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration episode artwork

EPISODE · Mar 28, 2026 · 22 MIN

Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration

from Daily Paper Cast · host Jingwen Liang, Gengyu Wang

🤗 Upvotes: 40 | cs.CV Authors: Danil Tokhchukov, Aysel Mirzoeva, Andrey Kuznetsov, Konstantin Sobolev Title: Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration Arxiv: http://arxiv.org/abs/2603.24800v1 Abstract: In this paper, we uncover the hidden potential of Diffusion Transformers (DiTs) to significantly enhance generative tasks. Through an in-depth analysis of the denoising process, we demonstrate that introducing a single learned scaling parameter can significantly improve the performance of DiT blocks. Building on this insight, we propose Calibri, a parameter-efficient approach that optimally calibrates DiT components to elevate generative quality. Calibri frames DiT calibration as a black-box reward optimization problem, which is efficiently solved using an evolutionary algorithm and modifies just ~100 parameters. Experimental results reveal that despite its lightweight design, Calibri consistently improves performance across various text-to-image models. Notably, Calibri also reduces the inference steps required for image generation, all while maintaining high-quality outputs.

Episode metadata supplied by the publisher feed · Published Mar 28, 2026

🤗 Upvotes: 40 | cs.CV Authors: Danil Tokhchukov, Aysel Mirzoeva, Andrey Kuznetsov, Konstantin Sobolev Title: Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration Arxiv: http://arxiv.org/abs/2603.24800v1 Abstract: In this paper, we uncover the hidden potential of Diffusion Transformers (DiTs) to significantly enhance generative tasks. Through an in-depth analysis of the denoising process, we demonstrate that introducing a single learned scaling parameter can significantly improve the performance of DiT blocks. Building on this insight, we propose Calibri, a parameter-efficient approach that optimally calibrates DiT components to elevate generative quality. Calibri frames DiT calibration as a black-box reward optimization problem, which is efficiently solved using an evolutionary algorithm and modifies just ~100 parameters. Experimental results reveal that despite its lightweight design, Calibri consistently improves performance across various text-to-image models. Notably, Calibri also reduces the inference steps required for image generation, all while maintaining high-quality outputs.

PodParley-generated summary based on available episode metadata and transcript content.

NOW PLAYING

Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration

0:00 22:30

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

Frequently Asked Questions

How long is this episode of Daily Paper Cast?

This episode is 22 minutes long.

When was this Daily Paper Cast episode published?

This episode was published on March 28, 2026.

What is this episode about?

🤗 Upvotes: 40 | cs.CV Authors: Danil Tokhchukov, Aysel Mirzoeva, Andrey Kuznetsov, Konstantin Sobolev Title: Calibri: Enhancing Diffusion Transformers via Parameter-Efficient...

Can I download this Daily Paper Cast episode?

Yes, you can download this episode by clicking the download button on the episode player, or subscribe to the podcast in your preferred podcast app for automatic downloads.
URL copied to clipboard!