Instructions to use Ryann829/Scone with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Ryann829/Scone with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Ryann829/Scone", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -16,7 +16,7 @@ library_name: transformers
|
|
| 16 |
<img src="assets/logo.png" alt="Scone" width="400"/>
|
| 17 |
</p>
|
| 18 |
<h3 align="center">
|
| 19 |
-
Scone: Bridging Composition and Distinction in Subject-Driven Image Generation
|
| 20 |
</h3>
|
| 21 |
|
| 22 |
<p align="center">
|
|
@@ -40,7 +40,7 @@ library_name: transformers
|
|
| 40 |
|
| 41 |
# 📢 News
|
| 42 |
|
| 43 |
-
|
| 44 |
- 2025.12.16: The [paper](https://arxiv.org/abs/2512.12675), [training code](https://github.com/Ryann-Ran/Scone?tab=readme-ov-file#-train), [inference and evaluation code](https://github.com/Ryann-Ran/Scone?tab=readme-ov-file#-inference-and-evaluation), [model weight](https://huggingface.co/Ryann829/Scone), [training data](https://huggingface.co/datasets/Ryann829/Scone-S2I-57K), [SconeEval benchmark](https://huggingface.co/datasets/Ryann829/SconeEval) are now released.
|
| 45 |
|
| 46 |
# 📖 Introduction
|
|
|
|
| 16 |
<img src="assets/logo.png" alt="Scone" width="400"/>
|
| 17 |
</p>
|
| 18 |
<h3 align="center">
|
| 19 |
+
[✨CVPR 2026 Highlight] Scone: Bridging Composition and Distinction in Subject-Driven</br> Image Generation via Unified Understanding-Generation Modeling
|
| 20 |
</h3>
|
| 21 |
|
| 22 |
<p align="center">
|
|
|
|
| 40 |
|
| 41 |
# 📢 News
|
| 42 |
|
| 43 |
+
- 2026.4.9: 🎉🎉 Our paper has been selected as a ✨**Highlight** at **CVPR 2026**.
|
| 44 |
- 2025.12.16: The [paper](https://arxiv.org/abs/2512.12675), [training code](https://github.com/Ryann-Ran/Scone?tab=readme-ov-file#-train), [inference and evaluation code](https://github.com/Ryann-Ran/Scone?tab=readme-ov-file#-inference-and-evaluation), [model weight](https://huggingface.co/Ryann829/Scone), [training data](https://huggingface.co/datasets/Ryann829/Scone-S2I-57K), [SconeEval benchmark](https://huggingface.co/datasets/Ryann829/SconeEval) are now released.
|
| 45 |
|
| 46 |
# 📖 Introduction
|