Align model card with the public ScaleNav code and project page

README.md

@@ -15,52 +15,58 @@ tags:
 
 **Scalable Internet Video Pretraining for Language-Conditioned Local Navigation**
 
-[Code](https://github.com/yangyihui0502/ScaleNav) · [Data](https://huggingface.co/datasets/Yangyihui/ScaleNav-Data) · [Project page](https://yangyihui0502.github.io/scalenav-review/)
+[Code](https://github.com/MachEmbodied/ScaleNav) · [Data](https://huggingface.co/datasets/Yangyihui/ScaleNav-Data) · [Project page](https://zju3dv.github.io/ScaleNav/)
 
-ScaleNav learns navigation priors from Internet videos through action-free
-pretraining, then predicts local actions and future observations conditioned on
-RGB observations and a navigation instruction.
+ScaleNav learns navigation priors from Internet egocentric videos through
+action-free pretraining and transfers them to action generation for
+language-conditioned local navigation. The goal is to reach a language-specified
+target visible in the initial observation and stop at the instructed location.
 
 ## Download
 
-From the ScaleNav code repository:
+Install ScaleNav using the [installation guide](https://github.com/MachEmbodied/ScaleNav#installation),
+then run from the code repository root. This model repository is public;
+authentication is optional:
 
 ```bash
-hf auth login
 python scripts/download_checkpoint.py --output checkpoints/scalenav
 ```
 
-The downloader uses a fixed Hugging Face revision and verifies each file's
-SHA-256. A [GitHub mirror](https://github.com/yangyihui0502/ScaleNav/releases/tag/scalenav-v1)
-is available with `--source github`.
+The downloader uses the revision and file checksums in the
+[download manifest](https://github.com/MachEmbodied/ScaleNav/blob/main/docs/checkpoint-download-manifest.json),
+resumes interrupted downloads and verifies SHA-256.
 
 To download directly with the Hugging Face CLI:
 
 ```bash
-hf download Yangyihui/ScaleNav --revision v1.0 --local-dir checkpoints/scalenav \
+hf download Yangyihui/ScaleNav --revision 8808b3589142b52e3bf0719aa0ecff8f86715a4a --local-dir checkpoints/scalenav \
   --include 'transformer/*' ar_model.pt train_meta.pt SHA256SUMS.json
 ```
 
 Download the Wan VAE, text encoder and tokenizer using the
-[installation and checkpoint guide](https://github.com/yangyihui0502/ScaleNav#checkpoints).
+[installation and checkpoint guide](https://github.com/MachEmbodied/ScaleNav#checkpoints).
 
 ## Inference
 
 Prepare the paired observation and its original instruction using the
-[data instructions](https://github.com/yangyihui0502/ScaleNav#data), then run:
+[single-image inference guide](https://github.com/MachEmbodied/ScaleNav/blob/main/docs/INFERENCE.md#single-image-inference), then run:
 
 ```bash
 awm check --checkpoint checkpoints/scalenav --sha256
 awm infer --checkpoint checkpoints/scalenav \
   --base-model checkpoints/Wan2.2-TI2V-5B \
   --image data/example/observation.png \
-  --prompt "$(cat data/example/instruction.txt)" --output outputs/example
+  --prompt "$(cat data/example/instruction.txt)" --seed 1065612570 \
+  --output outputs/example
 ```
 
-This produces a nine-step action trajectory and a generated video. Actions are
-per-frame forward/right/clockwise-yaw displacements in metres/metres/radians.
+This saves the nine-step action trajectory as `observation_trajectory.npy` and
+a trajectory visualization as `observation_trajectory.png`. Add `--decode-video`
+to also save nine predicted RGB frames as `observation_generated_raw.mp4`.
+Actions are per-frame forward/right/clockwise-yaw displacements in
+metres/metres/radians.
 Continuous navigation uses incoming camera frames through
-[`awm serve`](https://github.com/yangyihui0502/ScaleNav/blob/main/docs/INFERENCE.md).
+[`awm serve`](https://github.com/MachEmbodied/ScaleNav/blob/main/docs/INFERENCE.md).
 
 ## Model settings
 
@@ -85,6 +91,11 @@ Continuous navigation uses incoming camera frames through
 | `SHA256SUMS.json` | File checksums |
 
 Total checkpoint size: **10.02 GB**. The separate Wan components occupy about
-14.2 GB. The code repository includes [training](https://github.com/yangyihui0502/ScaleNav/blob/main/docs/TRAINING.md),
-[evaluation](https://github.com/yangyihui0502/ScaleNav/blob/main/docs/EVALUATION.md)
-and [trajectory visualization](https://github.com/yangyihui0502/ScaleNav#inference).
+14.2 GB. The code repository includes [training](https://github.com/MachEmbodied/ScaleNav/blob/main/docs/TRAINING.md),
+[evaluation](https://github.com/MachEmbodied/ScaleNav/blob/main/docs/EVALUATION.md)
+and [trajectory visualization](https://github.com/MachEmbodied/ScaleNav/blob/main/docs/INFERENCE.md#outputs).
+
+The released checkpoint is the final joint-fine-tuning checkpoint at step 6500.
+It contains model weights and deployment metadata. Optimizer, scheduler and RNG
+state for exact training resumption are not included. The Wan VAE, text encoder
+and tokenizer are downloaded separately.