You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Right now, we include first FRI layer into the FRI proof. This is redundant as the verifier can compute the values for the first FRI layer directly from trace and constraint queries. Removing first FRI layer from the proof should reduce of proof size roughly by 15%.
However, we can do this only when perfect zero knowledge is not required. When we implement ability to turn on perfect zero knowledge (via #9), we would need to include first FRI layer queries into the FRI proof. So, it probably makes sense to address this issue once #9 has been addressed.
The text was updated successfully, but these errors were encountered:
Right now, we include first FRI layer into the FRI proof. This is redundant as the verifier can compute the values for the first FRI layer directly from trace and constraint queries. Removing first FRI layer from the proof should reduce of proof size roughly by 15%.
However, we can do this only when perfect zero knowledge is not required. When we implement ability to turn on perfect zero knowledge (via #9), we would need to include first FRI layer queries into the FRI proof. So, it probably makes sense to address this issue once #9 has been addressed.
The text was updated successfully, but these errors were encountered: