Verifiable Fine-Tuning Secures Large Language Models with Zero-Knowledge Proofs
zkLoRA is a new framework that cryptographically verifies LLM fine-tuning correctness without revealing model weights, unlocking private, auditable AI.
Linear-Time Field-Agnostic SNARKs Unlock Massively Scalable Verifiable Computation
Brakedown introduces a practical linear-time encodable code, enabling the first O(N) SNARK prover, fundamentally scaling verifiable computation and ZK-Rollups.
Field-Agnostic Polynomial Commitments Unlock Fast, Universal Zero-Knowledge Proofs
BaseFold generalizes FRI, introducing foldable codes to create a field-agnostic polynomial commitment scheme with superior prover and verifier efficiency.
Sublinear Memory Zero-Knowledge Proofs Democratize Verifiable Computation
Introducing the first ZKP system with memory scaling to the square-root of computation size, this breakthrough enables privacy-preserving verification on edge devices.
