Zero-Knowledge Proof of Training Secures Federated Learning Consensus
ZKPoT uses zk-SNARKs to verify model contributions privately, eliminating the trade-off between decentralized AI privacy and consensus efficiency.
Zero-Knowledge Proof of Training Secures Private Decentralized AI Consensus
ZKPoT, a novel zk-SNARK-based consensus, cryptographically validates decentralized AI model contributions, eliminating privacy risks and scaling efficiency.
Dual-Proof Rollups Enable Configurable Cost-Finality Trade-Offs
This research pioneers a dual-proof rollup system, integrating ZK-STARKs and TEEs to deliver configurable finality and flexible security-cost trade-offs for Layer 2 solutions.
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.
GPU Bottlenecks Hinder Zero-Knowledge Proof Scalability and Adoption
This research identifies Number-Theoretic Transform as the primary GPU bottleneck for Zero-Knowledge Proofs, proposing architectural and tuning solutions to unlock verifiable computing at scale.
Achieving Statistical Non-Malleable Zero-Knowledge in Four Rounds
A novel four-round zero-knowledge argument achieves statistical non-malleability, advancing cryptographic proof systems beyond computational security.
Verifiable Data Aggregation Secures Decentralized Oracle Networks
A novel framework integrates cryptographic proofs with oracle aggregation, ensuring off-chain data integrity for robust smart contract execution.
ZKPoT: Private, Efficient Consensus for Federated Blockchain Learning
A novel Zero-Knowledge Proof of Training consensus mechanism secures federated learning, validating model contributions privately and efficiently on blockchains.
Sublinear-Space Zero-Knowledge Proving for Resource-Constrained Devices
A novel sublinear-space zero-knowledge prover reframes proof generation as tree evaluation, enabling efficient on-device verifiable computation for widespread adoption.
