Upload README.md with huggingface_hub

README.md

@@ -23,16 +23,18 @@ and German**. It is trained on **RefRad2D**, a large‑scale bilingual corpus de
 clinical routine with grounding labels generated automatically (TotalSegmentator +
 LLM‑based keyword extraction); the clinical dataset itself is **not** released.
 
-This repository holds the released checkpoints. Code, training, and evaluation:
-**https://github.com/lmb-freiburg/RadGrounder**
+This repository holds the released checkpoints.
+
+- 🌐 **Project page:** https://radgrounder.github.io
+- 💻 **Code, training, and evaluation:** https://github.com/lmb-freiburg/RadGrounder
 
 ## Contents
 
 | Subfolder | What it is | Grounding |
 |---|---|---|
-| [`detection/`](detection) | RadGrounder, token‑based bounding‑box grounding | boxes generated as text tokens |
-| [`segmentation/`](segmentation) | RadGrounder + lightweight mask decoder | `<seg>` spans → binary masks |
-| [`siglip/`](siglip) | the fine‑tuned SigLIP vision encoder (`.ckpt`) | — (only needed to **train** new models; the two checkpoints above already embed their encoder) |
+| [`detection/`](https://huggingface.co/lmb-freiburg/radgrounder/tree/main/detection) | RadGrounder, token‑based bounding‑box grounding | boxes generated as text tokens |
+| [`segmentation/`](https://huggingface.co/lmb-freiburg/radgrounder/tree/main/segmentation) | RadGrounder + lightweight mask decoder | `<seg>` spans → binary masks |
+| [`siglip/`](https://huggingface.co/lmb-freiburg/radgrounder/tree/main/siglip) | the fine‑tuned SigLIP vision encoder (`.ckpt`) | — (only needed to **train** new models; the two checkpoints above already embed their encoder) |
 
 Both checkpoints use a **frozen fine‑tuned SigLIP** vision encoder. Each model folder also
 contains its `training_config.json` (the exact training recipe).