Zero-Knowledge Proofs Verifiably Secure Large Language Model Inference
A novel ZKP system, zkLLM, enables the efficient, private verification of 13-billion-parameter LLM outputs, securing AI integrity and intellectual property.
New Lookup Argument Achieves Optimal Commitment Size for Universal ZK Circuits
Lasso introduces a sparse multilinear polynomial commitment scheme to make non-arithmetic ZK operations linear, unlocking the lookup singularity.
Lasso: Lookup Arguments Unlock Efficient Zero-Knowledge Computation
Lasso introduces a novel lookup argument that significantly optimizes zero-knowledge proofs by enabling efficient commitment to small field elements, transforming complex computations into succinct lookups.