Update technical_report_semantic.md

technical_report_semantic.md

@@ -8,8 +8,7 @@ and the decoder reconstructs via single-step VP diffusion.
 
 This checkpoint is trained with DINOv2 semantic alignment and variance
 expansion regularization. The name is a nod to DRA (Page et al., 2026) whose
-disentangled representation alignment approach inspired the semantic alignment
-method used here.
+disentangled representation alignment approach we largely follow here.
 
 ## Contents