paper-with-me

홈 › Papers

Dolfin: Diffusion Layout Transformers without Autoencoder

2023-10-25 · Yilin Wang, Zeyuan Chen, Liangjun Zhong, Zheng Ding, Zhizhou Sha, Zhuowen Tu

In this paper, we introduce a novel generative model, Diffusion Layout Transformers without Autoencoder (Dolfin), which significantly improves the modeling capability with reduced complexity compared to existing methods. Dolfin employs a Transformer-based diffusion process to model layout generation. In addition to an efficient bi-directional (non-causal joint) sequence representation, we further propose an autoregressive diffusion model (Dolfin-AR) that is especially adept at capturing rich semantic correlations for the neighboring objects, such as alignment, size, and overlap. When evaluated against standard generative layout benchmarks, Dolfin notably improves performance across various metrics (fid, alignment, overlap, MaxIoU and DocSim scores), enhancing transparency and interoperability in the process. Moreover, Dolfin's applications extend beyond layout generation, making it suitable for modeling geometric structures, such as line segments. Our experiments present both qualitative and quantitative results to demonstrate the advantages of Dolfin.

📄 PDF Abstract BibTeX arXiv:2310.16305

Code (0)

등록된 구현이 없습니다.

Tasks

Layout Generation

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

DOLFIN: Balancing Stability and Plasticity in Federated Continual Learning

2025-10-15 · Omayma Moussadek, Riccardo Salami, Simone Calderara arxiv

Federated continual learning (FCL) enables models to learn new tasks across multiple distributed clients, protecting privacy and without forgetting previously acquired knowledge. However, current methods face challenges …

Incremental LearningContinual Learning

Diffusion Transformers with Representation Autoencoders

2025-10-13 · Boyang Zheng, Nanye Ma, Shengbang Tong, Saining Xie arxiv

Latent generative modeling, where a pretrained autoencoder maps pixels into a latent space for the diffusion process, has become the standard strategy for Diffusion Transformers (DiT); however, the autoencoder component …

Image Generation

DoLFIn: Distributions over Latent Features for Interpretability

2020-11-10 · COLING 2020 8 · Phong Le, Willem Zuidema

Interpreting the inner workings of neural models is a key step in ensuring the robustness and trustworthiness of the models, but work on neural network interpretability typically faces a trade-off: either the models are …

text-classificationText Classification

Denoising Diffusion Autoencoders are Unified Self-supervised Learners

2023-03-17 · ICCV 2023 1 · Weilai Xiang, Hongyu Yang, Di Huang, Yunhong Wang

Inspired by recent advances in diffusion models, which are reminiscent of denoising autoencoders, we investigate whether they can acquire discriminative representations for classification via generative pre-training. Thi…

Contrastive LearningDenoisingImage GenerationLinear evaluation+2

SRA 2: Variational Autoencoder Self-Representation Alignment for Efficient Diffusion Training

2026-01-25 · Mengmeng Wang, Dengyang Jiang, Liuzhuozheng Li, Yucheng Lin 외 arxiv

Denoising-based diffusion transformers, despite their strong generation performance, suffer from inefficient training convergence. Existing methods addressing this issue, such as REPA (relying on external representation …