Update README.md

README.md

@@ -327,4 +327,4 @@ Training Scope	Only LoRA weights updated; main model remains fixed
 
 This approach enables self-corrective, explainable, and meta-aware learning, pushing beyond standard RLHF and toward autonomous reasoning agents.
 
-![GMPo Diagram](LiberalMind_v1.5/kl_critic_plot.png)
+![GMPo Diagram](https://huggingface.co/liberalusa/liberalmind_bin/blob/main/kl_critic_plot.png)