Research map / Image generation / Fast diffusion sampling
DiffInfinite: Large Mask-Image Synthesis via Parallel Random Patch Diffusion in Histopathology
Marco Aversa, Gabriel Nobis, Miriam Hägele, Kai Standvoss, Mihaela Chirica, Roderick Murray-Smith, Ahmed Alaa, Lukas Ruff, Daniela Ivanova, Wojciech Samek, Frederick Klauschen, Bruno Sanguinetti and 1 more
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| train classifier on synthetic data only (ablation) | classifier · IH2 classification | balanced accuracy: 0.733 -> 0.753 | Improved |
| train classifier on synthetic data only (ablation) | classifier · IH3 classification | balanced accuracy: 0.598 -> 0.699 | Improved |
| train classifier on synthetic data only (ablation) | classifier · IH1 classification (in-distribution) | balanced accuracy: 0.846 -> 0.747 | Hurt |
| train classifier on synthetic data only (ablation) | classifier · NCT-100K | balanced accuracy: 0.857 -> 0.796 | Hurt |
| train classifier on synthetic data only (ablation) | classifier · CRC-7K | balanced accuracy: 0.822 -> 0.753 | Hurt |
| train classifier on synthetic data only (ablation) | classifier · PCam-327K | balanced accuracy: 0.628 -> 0.628 | No clear effect |
| augment real data with synthetic data (ablation) | classifier · IH1 | balanced accuracy: 0.846 -> 0.852 | Improved |
| augment real data with synthetic data (ablation) | classifier · IH3 | balanced accuracy: 0.598 -> 0.637 | Improved |
| augment real data with synthetic data (ablation) | classifier · PCam-327K | balanced accuracy: 0.628 -> 0.641 | Improved |
| augment real data with synthetic data (ablation) | classifier · NCT-100K | balanced accuracy: 0.857 -> 0.847 | No clear effect |
| train segmenter on synthetic data only (ablation) | segmenter · IH2 segmentation | F1: 0.614 -> 0.471 | Hurt |
| augment real data with synthetic data (ablation) | segmenter · IH2 segmentation | F1: 0.614 -> 0.710 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DiffCollage — DiffCollage: Parallel Generation of Large Content With Diffusion Models (CVPR 2023) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Semantic and Visual Crop-Guided Diffusion Models for Heterogeneous Tissue Synthesis in Histopathology (NeurIPS 2025 · similarity 0.89)
- Dynamic Entity-Masked Graph Diffusion Model for Histopathology Image Representation Learning (AAAI 2025 · similarity 0.88)
- STREAM: Stochastic Riemannian Flow Matching with Anisotropic Decoder for Digital Histopathology Image Generation (NeurIPS 2026 · similarity 0.88)
- Beyond ViT Tokens: Masked-Diffusion Pretrained Convolutional Pathology Foundation Model for Cell-Level Dense Prediction (NeurIPS 2026 · similarity 0.88)
- ZoomLDM: Latent Diffusion Model for Multi-scale Image Generation (CVPR 2025 · similarity 0.87)
- L-Diffusion: Laplace Diffusion for Efficient Pathology Image Segmentation (ICML 2025 · similarity 0.86)
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.
