LayerSync: Self-aligning Intermediate Layers

Authors: Yasaman Haghighi*, Bastien van Delft*, Mariam Hassan, Alexandre Alahi
Affiliation: École Polytechnique Fédérale de Lausanne (EPFL)

---

📄 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 human motion generation. LayerSync consistently enhances the generation quality and the training efficiency. For example, it speeds up the training of flow-based transformer by over 8.75x on the ImageNet dataset and improves the generation quality by 23.6%.

---

🔽 Available Checkpoints

This repository contains pretrained model checkpoints:

• checkpoint.pt: Trained model after 800 epochs
• representation.pt: Checkpoint used for representation quality evaluation experiments

---

🚀 Usage

Download the checkpoints and use them with the LayerSync codebase.

---

📚 Citation

@misc{haghighi2025layersyncselfaligningintermediatelayers,
      title={LayerSync: Self-aligning Intermediate Layers}, 
      author={Yasaman Haghighi and Bastien van Delft and Mariam Hassan and Alexandre Alahi},
      year={2025},
      eprint={2510.12581},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2510.12581}, 
}

---

🔗 Links

• Paper: arXiv:2510.12581
• Project Page: vita-epfl.github.io/LayerSync.io
• Code: github.com/vita-epfl/LayerSync