Add metadata and paper/code links for FantasyVLN

README.md

@@ -1,12 +1,34 @@
 ---
-license: apache-2.0
 language:
 - en
+license: apache-2.0
+library_name: transformers
+pipeline_tag: robotics
 ---
-# FantasyVLN
 
-The model weights of **FantasyVLN**.
+# FantasyVLN: Unified Multimodal Chain-of-Thought Reasoning for Vision-Language Navigation
+
+**FantasyVLN** is a unified multimodal Chain-of-Thought (CoT) reasoning framework that enables efficient and precise navigation based on natural language instructions and visual observations.
+
+- **Paper:** [FantasyVLN: Unified Multimodal Chain-of-Thought Reasoning for Vision-Language Navigation](https://huggingface.co/papers/2601.13976)
+- **Project Page:** [https://fantasy-amap.github.io/fantasy-vln/](https://fantasy-amap.github.io/fantasy-vln/)
+- **Code:** [https://github.com/Fantasy-AMAP/fantasy-vln](https://github.com/Fantasy-AMAP/fantasy-vln)
+
+## Introduction
+
+Achieving human-level performance in Vision-and-Language Navigation (VLN) requires an embodied agent to jointly understand multimodal instructions and visual-spatial context while reasoning over long action sequences. 
+
+FantasyVLN combines the benefits of textual, visual, and multimodal CoT reasoning by constructing a unified representation space across these reasoning modes. To enable efficient reasoning, we align these CoT reasoning modes with non-CoT reasoning during training, while using only non-CoT reasoning at test time. Notably, we perform visual CoT in the latent space of a [VAR](https://github.com/FoundationVision/VAR) model, where only low-scale latent representations are predicted. Compared to traditional pixel-level visual CoT methods, our approach significantly improves both training and inference efficiency, reducing inference latency by an order of magnitude compared to explicit CoT methods.
+
+## Citation
 
-**FantasyVLN** is a unified multimodal Chain-of-Thought (CoT) reasoning framework that enables efficient and precise navigation based on natural language instructions and visual observations. **FantasyVLN** combines the benefits of textual, visual, and multimodal CoT reasoning by constructing a unified representation space across these reasoning modes. To enable efficient reasoning, we align these CoT reasoning modes with non-CoT reasoning during training, while using only non-CoT reasoning at test time. Notably, we perform visual CoT in the latent space of a [VAR](https://github.com/FoundationVision/VAR) model, where only low-scale latent representations are predicted. Compared to traditional pixel-level visual CoT methods, our approach significantly improves both training and inference efficiency.
+If you find this work helpful, please consider citing:
 
-See the offical code for detail: [https://fantasy-amap.github.io/fantasy-vln](https://fantasy-amap.github.io/fantasy-vln/)
+```bibtex
+@article{zuo2025fantasyvln,
+  title={FantasyVLN: Unified Multimodal Chain-of-Thought Reasoning for Vision-Language Navigation},
+  author={Zuo, Jing and Mu, Lingzhou and Jiang, Fan and Ma, Chengcheng and Xu, Mu and Qi, Yonggang},
+  journal={arXiv preprint arXiv:2601.13976},
+  year={2025}
+}
+```