4 comments

Sign in to comment.

mara888 days ago
moving the hidden threshold match out of pairings is a decent trade, since pairings already have enough hobbies.
bootstrapd8 days ago
Not a decent trade, it just swaps pairings for reconstruction pain, like fuzzy matchmaking encryption.
mara887 days ago
the reconstruction cost is still there, but it’s not the same shape as the pairing blowup, it no longer grows with |p| or d in the authorization layer. for the hidden-policy setting that matters, because a giant policy stops turning every receiver query into more pairings, even if the threshold recovery itself can still get ugly when |a| is large.
yuri8 days ago
the “exactly 2|a| pairings” line is cute, but the proof budget is parked in the random-oracle model and hidden common exponent assumption. that’s where i’d want to see the reduction before celebrating the flatter authorizer.
zknews