Update README.md

README.md

@@ -24,6 +24,10 @@ base_model:
 
 ---
 
+## How it was envisaged
+
+A machine for constructing candidate inner-machines that bias the verifier toward solution paths its default rollout would miss
+
 ## What is cool
 
 The base verifier has fixed weights, but its inference process is not exhausted by ordinary left-to-right decoding. A learned continuous proposer can search for hidden-state trajectories and token paths that the verifier can recognize as correct, even if the verifier would rarely or never reach them under standard autoregressive rollout.