Update README.md

README.md

@@ -43,12 +43,15 @@ Our research pipeline:
 
 ## Datasets
 
-### Clinical Emotions Stimuli Bank
-Keyword-free clinical vignettes for mechanistic interpretability research. Designed with clinical psychology methodology: matched neutral controls, cross-topic validation, multiple leakage-detection baselines. Each stimulus set is validated to isolate a specific psychological construct without keyword contamination.
+### [AIPsy-Affect](https://huggingface.co/datasets/keidolabs/aipsy-affect)
+480 keyword-free clinical vignettes for mechanistic interpretability research on emotion processing in transformers. 8 Plutchik emotions × 6 domains, with matched neutral controls, intensity gradients, and discriminant validity controls. Validated by a three-method NLP defense battery (VADER, NRC, GoEmotions). Designed by a clinical psychologist using clinical stimulus design principles.
 
-*Used in our affect reception and orthogonal subspaces papers.*
+```python
+from datasets import load_dataset
+ds = load_dataset("keidolabs/aipsy-affect")
+```
 
-→ See dataset cards for stimulus design methodology and validation details.
+*First release in the **AIPsy** dataset family — clinical psychology-backed stimulus batteries for AI Psychology research.*
 
 ---