Security Lunch ☀️ Ed. — Wednesday, August 5th, 2026, 12:00 pm @ CoDa E160
CoBBl: proof-oriented compilation for SNARK
Kunming Jiang
Can't make it in person? Join us on
zoom.
See our past & upcoming events on our
website!
Abstract:
Succinct Non-Interactive Arguments of Knowledge, or SNARKs, allow a prover to efficiently convince a verifier that a computation was performed correctly. They are increasingly used in settings such as cloud computing and blockchains. Most existing SNARK research
focuses either on improving proof protocols or on designing program representations that are more compatible with them. CoBBl instead introduces an execution model that more closely ties program representation with the underlying proof system. CoBBl uses a
block-based execution model that can be instantiated over a wide range of existing proof protocols. By exposing proof-system-specific structure, it avoids inefficiencies found in conventional SNARK execution models and enables the compiler to apply proof-oriented
optimizations more broadly than previous systems. Together, these techniques reduce proving time by 3 - 5x compared with state-of-the-art SNARK systems.
The work is a collaboration among Kunming, Fraser, and Riad.
Bio:
Kunming Jiang is a fourth-year PhD student at Carnegie Mellon University advised by Fraser Brown and Riad Wahby.