Link reference data generation pipeline

README.md

@@ -19,16 +19,13 @@ weights associated
 with [*CSI-MAE: A Masked Autoencoder-based Channel Foundation Model*](https://arxiv.org/abs/2601.03789).
 
 - Code: [GREAT-ISAC/CSI-MAE](https://github.com/GREAT-ISAC/CSI-MAE)
-
-Released by **[GREAT Wireless AI](https://github.com/GREAT-ISAC)**. The
-`GREAT-ISAC` GitHub handle is retained for continuity with existing papers and
-community links.
+- Reference data generation: [GREAT-ISAC/Channel-Simulation-Data](https://github.com/GREAT-ISAC/Channel-Simulation-Data)
 - Paper: [arXiv:2601.03789](https://arxiv.org/abs/2601.03789)
 
 ## Weight files
 
 The release contains epoch-300 Base and Large checkpoints pretrained on the
-simulated Sionna/3GPP channel corpus. These are not the separate DeepMIMO
+simulated Sionna/3GPP channel data. These are not the separate DeepMIMO
 experimental checkpoints. None of the files contains an optimizer, AMP scaler,
 epoch/resume state, training data, or a downstream task head.
 
@@ -39,7 +36,7 @@ epoch/resume state, training data, or a downstream task head.
 | `csi-mae-large.safetensors` | Large model-only pre-trained weights | Recommended Large weight for safe standalone loading and feature extraction |
 | `csi-mae-large.pth` | Large model-only PyTorch checkpoint with a `model` key | Large compatibility weight for existing scripts |
 
-These are foundation-model pre-training weights, not final checkpoints for
+These are channel-foundation-model pre-training weights, not final checkpoints for
 positioning, channel feedback, or channel extrapolation. The corresponding
 downstream architecture must be initialized from these weights and then
 adapted or fine-tuned. Exact SHA-256 values are recorded in `manifest.json`.
@@ -112,9 +109,10 @@ python load_pretrained.py \
 The Base and Large models were pre-trained on simulated Sionna/3GPP CSI. The
 separate DeepMIMO experiments are not the source of these published weights.
 Generated training arrays are not included in this model repository. A
-**reproducible data-generation pipeline** will be released separately with the
-simulator configuration, preprocessing, normalization statistics, and split
-definitions.
+**reproducible, model-compatible reference data-generation pipeline** is
+available in [Channel Simulation Data](https://github.com/GREAT-ISAC/Channel-Simulation-Data).
+Its committed Sionna configuration is a runnable reference example; it does
+not reconstruct the complete checkpoint training data.
 
 ## Intended use
 
@@ -137,7 +135,7 @@ definitions.
 
 ## License
 
-The original CSI-MAE code, these model weights, and the accompanying original
+The original CSI-MAE code, these model weights, and the repository-owned
 data-generation scripts are released under the
 [Creative Commons Attribution-NonCommercial 4.0 International](https://creativecommons.org/licenses/by-nc/4.0/)
 license (**CC BY-NC 4.0**). Attribution is required and commercial use is not