Move metadata to YAML header and add paper links

README.md

@@ -1,18 +1,20 @@
-# [CVPR-26 🎉] SeeThrough3D: Occlusion Aware 3D Control in Text-to-Image Generation
-
-<!-- ---
-pipeline_tag: occlusion aware 3d control in text-to-image generation    # Specify the task
-library_name: diffusers       # Specify the library
-language:
-  - en                           # List languages your model supports
-license: apache-2.0              # Specify a license
+---
+license: apache-2.0
+library_name: diffusers
+pipeline_tag: text-to-image
+base_model: black-forest-labs/FLUX.1-dev
 datasets:
-  - va1bhavagrawal/seethrough3d_dataset            # List datasets used for training
-base_model: black-forest-labs/FLUX.1-dev  # If applicable (your model is a fine-tune, quantized, merged version of another model)
-tags:                            # Add extra tags which would make the repo searchable using the tag
-  - controllable text-to-image generation
-  - diffusion models 
---- -->
+- va1bhavagrawa1/seethrough3d-data
+language:
+- en
+tags:
+- controllable text-to-image generation
+- diffusion models
+- 3D layout control
+- occlusion reasoning
+---
+
+# [CVPR-26 🎉] SeeThrough3D: Occlusion Aware 3D Control in Text-to-Image Generation
 
 [![arXiv](https://img.shields.io/badge/arXiv-2602.23359-b31b1b.svg)](https://arxiv.org/abs/2602.23359)
 [![Project Page](https://img.shields.io/badge/Project-Page-blue.svg)](https://seethrough3d.github.io)