Upload README.md with huggingface_hub

README.md

@@ -139,7 +139,7 @@ for idx in top_k:
 
 ### Masked reconstruction heatmap (explainability)
 
-The model was pretrained to reconstruct masked token geometry statistics from surrounding context — this gives you a built-in per-region attribution map. High reconstruction error regions are where the model's learned geometric prior considers the input novel or surprising. See the [inference demo repo](https://github.com/simd-ai/shape-backend-v2) for a full implementation with 3D visualization.
+The model was pretrained to reconstruct masked token geometry statistics from surrounding context — this gives you a built-in per-region attribution map. High reconstruction error regions are where the model's learned geometric prior considers the input novel or surprising.
 
 ## Training Data
 
@@ -183,8 +183,8 @@ Train / val split is deterministic (58,069 train / 2,983 val, 4.89%) using md5 h
 ## Citation
 
 ```bibtex
-@software{simd_ai_shape_foundation_2026,
-  author = {SIMD AI},
+@software{notelink_shape_foundation_2026,
+  author = {{Notelink LLC}},
   title  = {Shape Foundation Model},
   year   = {2026},
   url    = {https://huggingface.co/simd-ai/shape-foundation-small-v3}
@@ -193,6 +193,4 @@ Train / val split is deterministic (58,069 train / 2,983 val, 4.89%) using md5 h
 
 ## Links
 
-- **Training code**: [github.com/simd-ai/shape-v2](https://github.com/simd-ai/shape-v2)
-- **Inference demo**: [github.com/simd-ai/shape-backend-v2](https://github.com/simd-ai/shape-backend-v2)
-- **Frontend demo**: [github.com/simd-ai/shape-frontend-v2](https://github.com/simd-ai/shape-frontend-v2)
+- **Demo**: [github.com/simd-ai/shape-v2](https://github.com/simd-ai/shape-v2)