Towards Aligned Layout Generation via Diffusion Model with Aesthetic Constraints

Authors: Jian Chen, Ruiyi Zhang, Yufan Zhou, Changyou Chen

ICLR 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiment results show that LACE produces high-quality layouts and outperforms existing state-of-the-art baselines.
Researcher Affiliation Collaboration University at Buffalo1, Adobe Research2, MBZUAI3
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code Yes 1Code is available at https://github.com/puar-playground/LACE
Open Datasets Yes We use two large-scale datasets for comparison. Pub Lay Net (Zhong et al., 2019) consists of 330K document layouts... Rico (Deka et al., 2017) consists of 72k user interfaces designs...
Dataset Splits Yes The refined Pub Lay Net and Rico datasets were then split into training, validation, and test sets containing 315, 757/16, 619/11, 142 and 35, 851/2, 109/4, 218 samples respectively.
Hardware Specification No The paper does not provide specific hardware details used for running its experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes The batch size is 256. We used a learning rate schedule that included a warmup phase followed by a half-cycle cosine decay. The initial learning rate is set to 0.001. ... The diffusion model employs a total of 1000 forward steps. For efficient generation, we use DDIM sampling with 100 steps.