Update README.md

README.md

@@ -153,7 +153,7 @@ world_points = predictions["world_points"][0]   # (S, H, W, 3)
 **Other Properties Related to Output:**
 - One 3D point, one depth value, one ray, and one depth-confidence value are predicted per input pixel in each image.
 - One set of camera parameters (intrinsics + extrinsics) is predicted per input image, expressed in the coordinate frame of the first view.
-- Cameras are recovered from the predicted ray maps following [Lin et al. 2025](https://arxiv.org/abs/2511.10647); a camera multi-layer perceptron (MLP) head provides a faster auxiliary alternative.
+- Cameras are recovered from the predicted ray maps following [Lin et al. 2025](https://github.com/bytedance-seed/depth-anything-3); a camera multi-layer perceptron (MLP) head provides a faster auxiliary alternative.
 
 Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA's hardware (e.g. GPU cores) and software frameworks (e.g., Compute Unified Device Architecture (CUDA) libraries), the model achieves faster training and inference times compared to CPU-only solutions. <br>