ZKPoT Secures Decentralized Machine Learning by Proving Training without Revealing Data
A new ZKPoT consensus uses zk-SNARKs to cryptographically verify decentralized AI model training performance while preserving data privacy.
Homogeneous Weight Functions Secure Multi-Resource Longest-Chain Consensus
A mathematical classification of resource-weighting functions secures longest-chain protocols, ensuring persistence against private double-spending attacks.
Revelation Mechanisms Enforce Truthful, Fork-Free Consensus in Proof-of-Stake
This mechanism design breakthrough uses revelation principles to create a unique, truthful equilibrium, fundamentally securing PoS against adversarial block proposal.
Unifying Consensus Resilience Bounds under Adversary Majority
This research systematizes 16 consensus models to resolve conflicting security bounds, demonstrating safety up to 99% adversarial stake.
Optimal Prover Time Succinct Zero-Knowledge Proofs Redefine Scalability
The Libra proof system achieves optimal linear prover time, solving the primary bottleneck of ZKPs to unlock practical, large-scale verifiable computation.
Active Block Producers Create Impossibility for Incentive-Compatible Fee Mechanisms
Formal analysis proves active block producers, driven by private MEV, fundamentally prevent simultaneous incentive-compatibility and welfare-maximization.
Plonky2-FRI Enables Scalable Zero-Knowledge Proof for Cryptographic Hashing Verification
This research integrates Plonky2 with FRI to generate succinct proofs for SHA-256 integrity, fundamentally decoupling computational work from verification cost.
Adaptive Byzantine Agreement Achieves Optimal Communication Complexity with Few Faults
A new Byzantine Agreement protocol achieves optimal $O(n+t cdot f)$ adaptive communication complexity, scaling cost by actual faults, not maximum potential faults.
Layered Commit-Reveal Protocol Secures Decentralized Randomness Beacons
Commit-Reveal Squared uses randomized reveal order and a hybrid architecture to cryptographically secure decentralized randomness, eliminating last-revealer bias.
