# LayerSync: Self-aligning Intermediate Layers

Source: [LayerSync: Self-aligning Intermediate Layers](<https://www.sota2.com/research/paper/68fa7c1b1e209131cea8c637>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-21T05:00:04.173Z

Authors: Yasaman Haghighi, Bastien van Delft, Mariam Hassan, Alexandre Alahi

Published: 2025-10-14T14:39:14.000Z

## Abstract

We propose LayerSync, a domain-agnostic approach for improving the generation quality and the training efficiency of diffusion models. Prior studies have highlighted the connection between the quality of generation and the representations learned by diffusion models, showing that external guidance on model intermediate representations accelerates training. We reconceptualize this paradigm by regularizing diffusion models with their own intermediate representations. Building on the observation that representation quality varies across diffusion model layers, we show that the most semantically rich representations can act as an intrinsic guidance for weaker ones, reducing the need for external supervision. Our approach, LayerSync, is a self-sufficient, plug-and-play regularizer term with no overhead on diffusion model training and generalizes beyond the visual domain to other modalities. LayerSync requires no pretrained models nor additional data. We extensively evaluate the method on image generation and demonstrate its applicability to other domains such as audio, video, and motion generation. We show that it consistently improves the generation quality and the training efficiency. For example, we speed up the training of flow-based transformer by over 8.75x on ImageNet dataset and improved the generation quality by 23.6%. The code is available at https://github.com/vita-epfl/LayerSync.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Image Generation on ImageNet 256x256](<https://www.sota2.com/research/sota/image-generation-on-imagenet-256x256.md>): Image Generation; ImageNet 256x256; IS: 265.3.

- [Class-conditional Image Generation on ImageNet 256x256](<https://www.sota2.com/research/sota/class-conditional-image-generation-on-imagenet-256x256.md>): Class-conditional Image Generation; ImageNet 256x256; Inception Score (IS): 285.2.

- [Image Generation on ImageNet 256x256 (val)](<https://www.sota2.com/research/sota/image-generation-on-imagenet-256x256-val.md>): Image Generation; ImageNet 256x256 (val); FID: 8.8.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2510.12581>)

[PDF](<https://arxiv.org/pdf/2510.12581.pdf>)
