Research map / Image generation / Autoregressive and latent generation
Generative Adversarial Transformers
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Replace StyleGAN2's single global latent modulation with GANsformer bipartite attention | GANsformer · CLEVR | FID: 16.05 -> 9.17 | Helped |
| Replace StyleGAN2's single global latent modulation with GANsformer bipartite attention | GANsformer · LSUN-Bedroom | FID: 11.53 -> 6.51 | Helped |
| Replace StyleGAN2's single global latent modulation with GANsformer bipartite attention | GANsformer · FFHQ | FID: 9.24 -> 7.42 | Helped |
| Replace StyleGAN2's single global latent modulation with GANsformer bipartite attention | GANsformer · Cityscapes | FID: 8.35 -> 5.76 | Helped |
| Use duplex rather than simplex attention (ablation) | GANsformer · CLEVR | FID: 10.26 -> 9.17 | Helped |
| Use duplex rather than simplex attention (ablation) | GANsformer · LSUN-Bedroom | FID: 8.56 -> 6.51 | Helped |
| Use duplex rather than simplex attention (ablation) | GANsformer · FFHQ | FID: 8.12 -> 7.42 | Helped |
| Use duplex rather than simplex attention (ablation) | GANsformer · Cityscapes | FID: 14.23 -> 5.76 | Helped |
| Apply attention in more generator layers rather than fewer layers (ablation) | GANsformer · CLEVR | FID | Helped |
Closest papers by meaning
- Compositional Transformers for Scene Generation (NeurIPS 2021 · similarity 0.93)
- TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up (NeurIPS 2021 · similarity 0.90)
- Scalable GANs with Transformers (ICML 2026 · similarity 0.90)
- StyleSwin: Transformer-Based GAN for High-Resolution Image Generation (CVPR 2022 · similarity 0.90)
- A Geometric Analysis of Deep Generative Image Models and Its Applications (ICLR 2021 · similarity 0.90)
- The GAN is dead; long live the GAN! A Modern GAN Baseline (NeurIPS 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.