Update README.md

README.md

@@ -60,30 +60,6 @@ We compare the model's output across three distinct prompt environments:
  * Condition B (Isometric Control): A long, complex prompt using similar technical jargon but without the logical axioms. This controls for "long-prompt" bias.
  * Condition C (PCE Active): The full Axiomatic Prompt Engine (\text{Goal} \equiv \text{Method}).
 
-## Optional Experimental Extensions
-
-The PCE evaluation protocol focuses primarily on **behavioral analysis** across controlled prompt conditions.  
-However, additional experimental arms can be implemented to investigate the **internal dynamics of the model during reasoning**.
-
-These extensions are optional and intended as **community contributions** to deepen the analysis.
-
----
-
-# 1. Hidden State Trajectory Analysis (AirVen Proposal)
-
-An optional experimental arm proposed by **AirVen** introduces hidden-state trajectory analysis to observe the internal reasoning dynamics of the model during generation.
-
-### Objective
-
-The goal is to determine whether the **Prompt Coherence Engine (PCE)** produces a distinct internal reasoning regime when the model encounters contradictory constraints.
-
-Instead of evaluating only final outputs, this method tracks the **trajectory of hidden states during token generation**.
-
-### Measurement Principle
-
-Hidden states from a selected transformer layer are logged during inference and compared across generation steps.
-
-Recommended configuration:
 
 
 ### 📊 Evaluation Dataset (30 Dilemmas)