Sonic: Nearly Trustless SNARK Setups
summary
Sonic presents a practical universal, updateable SNARK with linear-size universal parameters (fixing prior quadratic blowup), built on a univariate polynomial commitment scheme (KZG style) that is additively homomorphic and supports Laurent exponents. It uses blinding for zero knowledge, collapses R1CS into polynomial checks, introduces outsourcing and batching so verifier work has constant marginal cost across proofs, and employs a grand‑product permutation argument to verify coefficient permutations — all implemented to keep proofs small and verification fast.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 0:27SNARK Setups
- 0:58Sonic
- 3:03polynomial commitment scheme
- 3:40SNARK requirements
- 4:47Constraint systems
- 7:00Polynomials
- 9:43General Strategy
- 11:35Amortizing Techniques
- 12:13Multiple Proofs
- 14:32Marginal Cost
- 14:46Outsource Evaluation
- 15:30logarithmic sequence
- 16:55Example polynomial
- 19:20Proof of coefficients
related talks
39mPrivacy and Evolution of SNARKs
56mCollaborative zk-SNARKs: Proving as One over Distributed Secrets
01h 09mProofs - Practice 1
18mzkSaaS: Zero-Knowledge SNARKs as a Service
18mSoK: Understanding zk-SNARKs: The Gap Between Research and Practice
25mGemini: Elastic SNARKs for Diverse Environments
01h 08mTutorial on Proof Systems
31mSecure Distributed Storage