We are pleased to announce our latest publication, “A Fully-Adaptive Threshold Partially-Oblivious PRF”, which will be presented at CRYPTO'25.

Abstract:

Oblivious Pseudorandom Functions (OPRFs) are fundamental cryptographic primitives essential for privacy-enhancing technologies such as private set intersection, oblivious keyword search, and password-based authentication protocols. We present the first fully adaptive, partially oblivious threshold pseudorandom function that supports proactive key refresh and provides composable security under the One-More Gap Diffie-Hellman assumption in the random oracle model.Our construction is secure with respect to a new ideal functionality for OPRFs that addresses three critical shortcomings of previous models - specifically, key refresh and non-verifiability issues that rendered them unrealizable. In addition, we identify a gap in a prior work’s proof of partial obliviousness and develop a novel proof technique to salvage their scheme.

Stay tuned for more details and discussions on this topic at CRYPTO!