Improve model card: Add robotics pipeline tag and canonical links

This PR enhances the model card for MolmoAct-7B-D by:

- Adding the `pipeline_tag: robotics` to the metadata, which helps users discover the model via the Hugging Face Hub's pipeline filters (e.g., at https://huggingface.co/models?pipeline_tag=robotics).
- Updating the paper link in the "Quick links" section to point to the Hugging Face Papers page ([https://huggingface.co/papers/2508.07917](https://huggingface.co/papers/2508.07917)) for consistency and improved discoverability within the Hub.
- Adding a direct link to the GitHub repository ([https://github.com/allenai/MolmoAct](https://github.com/allenai/MolmoAct)) in the "Quick links" section for easier access to the codebase.

These changes will help researchers and practitioners more easily find and understand the model's capabilities and resources.

README.md

@@ -1,11 +1,12 @@
 ---
-license: apache-2.0
-language:
-- en
 base_model:
 - Qwen/Qwen2.5-7B
 - google/siglip2-so400m-patch14-384
+language:
+- en
 library_name: transformers
+license: apache-2.0
+pipeline_tag: robotics
 tags:
 - molmoact
 - molmo
@@ -21,7 +22,7 @@ tags:
 # MolmoAct 7B-D
 
 MolmoAct is a fully open-source action reasoning model for robotic manipulation developed by the Allen Institute for AI. MolmoAct is trained on a subset of OXE and MolmoAct Dataset, a dataset with 10k high-quality trajectories of a single-arm Franka robot performing 93 unique manipulation tasks in both home and tabletop environments. It has state-of-the-art performance among vision-language-action models on multiple benchmarks while being fully open-source. You can find all models in the MolmoAct family [here](https://huggingface.co/collections/allenai/molmoact-689697591a3936fba38174d7).
-**Learn more about MolmoAct** in our announcement [blog post](https://allenai.org/blog/molmoact) or the [paper](https://huggingface.co/allenai/MolmoAct-7B-D-0812/blob/main/MolmoAct_Technical_Report.pdf).
+**Learn more about MolmoAct** in our announcement [blog post](https://allenai.org/blog/molmoact) or the [paper](https://huggingface.co/papers/2508.07917).
 
 **MolmoAct 7B-D** is based on [Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) and uses [SigLip2](https://huggingface.co/google/siglip2-so400m-patch14-384) as the vision backbone, which is initialized using Molmo's pre-training approach. It is first pre-trained on MolmoAct's [Pre-training Mixture](https://huggingface.co/datasets/allenai/MolmoAct-Pretraining-Mixture), and then mid-trained on [MolmoAct Dataset](https://huggingface.co/datasets/allenai/MolmoAct-Midtraining-Mixture). This model is intended to be used for downstream post-training.
 
@@ -30,7 +31,8 @@ This checkpoint is a **preview** of the MolmoAct release. All artifacts used in
 Quick links:
 - 📂 [All Models](https://huggingface.co/collections/allenai/molmoact-689697591a3936fba38174d7)
 - 📂 [All Data](https://huggingface.co/collections/allenai/molmoact-data-mixture-6897e583e13b6c2cf3ea2b80)
-- 📃 [Paper](https://arxiv.org/pdf/2508.07917)
+- 📄 [Paper](https://huggingface.co/papers/2508.07917)
+- 💻 [GitHub Repository](https://github.com/allenai/MolmoAct)
 - 🎥 [Blog Post](https://allenai.org/blog/molmoact)
 - 🎥 [Video](https://youtu.be/-_wag1X25OE?si=Xi_kUaJTmcQBx1f6)