Stochastic Networks Enable Logarithmic Broadcast and Consensus Resilience
New research reveals that distributed consensus in dynamic, unreliable networks can achieve logarithmic time complexity by embracing stochasticity, overcoming pessimistic deterministic limitations.
Asymmetric Trust Redefines Distributed Fault Tolerance
This research introduces asymmetric Byzantine quorum systems, enabling subjective trust models to secure distributed protocols and consensus mechanisms.
Off-Chain Mechanisms Unlock Scalable Relational Blockchain Databases
GriDB pioneers off-chain cross-shard data services, leveraging succinct proofs and delegation to overcome blockchain database scalability limitations.
ZKPoT Secures Federated Learning, Ensuring Privacy and Efficiency in Decentralized Systems
A novel Zero-Knowledge Proof of Training (ZKPoT) consensus validates model performance privately, enabling scalable, secure federated learning.
Pod: Optimal-Latency, Censorship-Free, Accountable Generalized Consensus Layer
A novel consensus primitive eliminates inter-replica communication, achieving physically optimal latency and enabling high-speed, accountable decentralized applications.
Self-Stabilizing Replicated State Machines Resist Byzantine and Recurring Transient Faults
This paper introduces the first protocol for repeated Byzantine agreement that integrates self-stabilization, enabling distributed systems to autonomously recover from both malicious and transient errors.
Hierarchical Consensus Enhances Blockchain Scalability and Fault Tolerance
A novel hierarchical consensus algorithm boosts blockchain transaction throughput and reduces latency by balancing workload across dynamic, multi-layered nodes.
Blockchain Epidemic Consensus Protocol Enables Scalable, Leaderless Agreement for Large Networks
BECP pioneers a scalable, leaderless epidemic consensus protocol for large blockchain networks, boosting throughput and efficiency while preserving decentralization.
Zero-Knowledge Proof of Training Secures Decentralized Federated Learning Consensus
ZKPoT uses zk-SNARKs to verify decentralized model accuracy without revealing private data, solving the efficiency-privacy trade-off in federated learning.
