Update the paper link in README.md.

README.md

@@ -17,7 +17,7 @@ CogACT is a new advanced VLA architecture derived from VLM. Unlike previous work
 
 All our [code](https://github.com/microsoft/CogACT), [pretrained model weights](https://huggingface.co/CogACT), are licensed under the MIT license.
 
-Please refer to our [project page](https://cogact.github.io/) and [paper](https://cogact.github.io/CogACT_paper.pdf) for more details.
+Please refer to our [project page](https://cogact.github.io/) and [paper](https://arxiv.org/abs/2411.19650) for more details.
 
 
 ## Model Summary
@@ -32,7 +32,7 @@ Please refer to our [project page](https://cogact.github.io/) and [paper](https:
   + **Action Model**: DiT-Large
 - **Pretraining Dataset:** A subset of [Open X-Embodiment](https://robotics-transformer-x.github.io/)
 - **Repository:** [https://github.com/microsoft/CogACT](https://github.com/microsoft/CogACT)
-- **Paper:** [CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation](https://cogact.github.io/CogACT_paper.pdf)
+- **Paper:** [CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation](https://arxiv.org/abs/2411.19650)
 - **Project Page:** [https://cogact.github.io/](https://cogact.github.io/)
 
 ## Uses