Verifiable Information Dispersal Decouples Finality from Asynchronous Data Availability
Asynchronous Verifiable Dispersal is a new primitive enabling optimal BFT latency by proving data dispersal before full reconstruction, accelerating consensus.
Formal Rewrites Safely Scale Byzantine Fault Tolerance Protocols Fivefold
A new formal model and rewrite methodology prove that decoupling and partitioning can safely achieve a fivefold throughput increase for BFT protocols.
Graded Dispersal Simplifies BFT Protocols Reducing Complexity and Communication Overhead
Foundational BFT protocols are simplified through Graded Dispersal, a new primitive that cuts communication complexity by 40% and reduces consensus rounds.
Erasure Coding Achieves Near-Optimal Byzantine Broadcast Communication Efficiency
Researchers deployed erasure-correcting codes and vector commitments to fragment messages, drastically reducing Byzantine Reliable Broadcast communication complexity to near-optimal bounds.
Decentralized Rollup Sequencers Using Set Consensus Ensure Full L2 Autonomy
Set Byzantine Consensus creates a decentralized arranger service, eliminating the sequencer bottleneck and enabling fully autonomous Layer 2 rollups.
Hashgraph Consensus Secures Multi-Model AI Reasoning, Eliminating LLM Hallucinations
Applying BFT-secure Hashgraph to LLM ensembles creates a novel, iterative consensus protocol that formally verifies model outputs, dramatically boosting AI reliability.
Graded Broadcast Protocol Advances Asynchronous BFT for Ultra-Low Latency and Throughput
Falcon's Graded Broadcast primitive allows asynchronous BFT protocols to bypass the costly agreement stage, fundamentally resolving the long-standing latency bottleneck in distributed consensus.
Concurrent BFT Decouples Throughput and Latency, Eliminating Censorship
A new asynchronous BFT protocol concurrently executes dissemination and agreement, resolving the throughput-latency tradeoff and guaranteeing censorship resistance.
Formally Synthesizing Secure Distributed Systems from Centralized Programs
This research unifies simulation-based security with compiler techniques to automatically generate provably secure distributed cryptographic applications.
