惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

U
Unit 42
罗磊的独立博客
T
Tailwind CSS Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Jina AI
Jina AI
V
V2EX
美团技术团队
阮一峰的网络日志
阮一峰的网络日志
酷 壳 – CoolShell
酷 壳 – CoolShell
月光博客
月光博客
量子位
MyScale Blog
MyScale Blog
G
Google Developers Blog
M
MIT News - Artificial intelligence
L
LangChain Blog
Microsoft Azure Blog
Microsoft Azure Blog
Recent Announcements
Recent Announcements
MongoDB | Blog
MongoDB | Blog
N
Netflix TechBlog - Medium
有赞技术团队
有赞技术团队
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
D
DataBreaches.Net
云风的 BLOG
云风的 BLOG
B
Blog

cs.DC updates on arXiv.org

DUAL-BLADE: Dual-Path NVMe-Direct KV-Cache Offloading for Edge LLM Inference Progressive Semantic Communication for Efficient Edge-Cloud Vision-Language Models Efficient, VRAM-Constrained xLM Inference on Clients Folding Tensor and Sequence Parallelism for Memory-Efficient Transformer Training & Inference DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Training AMMA: A Multi-Chiplet Memory-Centric Architecture for Low-Latency 1M Context Attention Serving RaMP: Runtime-Aware Megakernel Polymorphism for Mixture-of-Experts Spark Policy Toolkit: Semantic Contracts and Scalable Execution for Policy Learning in Spark Internet of Everything in the 6G Era: Paradigms, Enablers, Potentials and Future Directions PolyKV: A Shared Asymmetrically-Compressed KV Cache Pool for Multi-Agent LLM Inference A Survey on Split Learning for LLM Fine-Tuning: Models, Systems, and Privacy Optimizations ITAS: A Multi-Agent Architecture for LLM-Based Intelligent Tutoring Latency and Cost of Multi-Agent Intelligent Tutoring at Scale TACO: Efficient Communication Compression of Intermediate Tensors for Scalable Tensor-Parallel LLM Training FreeScale: Distributed Training for Sequence Recommendation Models with Minimal Scaling Cost CommFuse: Hiding Tail Latency via Communication Decomposition and Fusion for Distributed LLM Training A Taxonomy and Resolution Strategy for Client-Level Disagreements in Federated Learning Usable Agent Discovery for Decentralized AI Systems Cloud to Edge: Benchmarking LLM Inference On Hardware-Accelerated Single-Board Computers Data-Free Contribution Estimation in Federated Learning using Gradient von Neumann Entropy Shard the Gradient, Scale the Model: Serverless Federated Aggregation via Gradient Partitioning Promoting Simple Agents: Ensemble Methods for Event-Log Prediction GraphLeap: Decoupling Graph Construction and Convolution for Vision GNN Acceleration on FPGA AGNT2: Autonomous Agent Economies on Interaction-Optimized Layer 2 Infrastructure FedSIR: Spectral Client Identification and Relabeling for Federated Learning with Noisy Labels Stream-CQSA: Avoiding Out-of-Memory in Attention Computation via Flexible Workload Scheduling A Delta-Aware Orchestration Framework for Scalable Multi-Agent Edge Computing Federated Learning over Blockchain-Enabled Cloud Infrastructure Optimal Routing for Federated Learning over Dynamic Satellite Networks: Tractable or Not? Sherpa.ai Privacy-Preserving Multi-Party Entity Alignment without Intersection Disclosure for Noisy Identifiers
A Tendermint Light Client
Sean Braithwaite, Ethan Buchman, Ismail Khoffi, Igor Konnov, Zar · 2020-10-14 · via cs.DC updates on arXiv.org

In Tendermint blockchains, the proof-of-stake mechanism and the underlying consensus algorithm entail a dynamic fault model that implies that the active validators (nodes that sign blocks) may change over time, and a quorum of these validators is assumed to be correct only for a limited period of time (called trusting period). The changes of the validator set are under control of the blockchain application, and are committed in every block. In order to check what is the state of the blockchain application at some height h, one needs to know the validator set at that height so that one can verify the corresponding digital signatures and hashes. A naive way of determining the validator set for height h requires one to: (i) download all blocks before h, (ii) verify blocks by checking digital signatures and hashes and (iii) execute the corresponding transactions so the changes in the validator sets are reproduced. This can potentially be very slow and computationally and data intensive. In this paper we formalize the dynamic fault model imposed by Tendermint, and describe a light client protocol that allows to check the state of the blockchain application that, in realistic settings, reduces significantly the amount of data needed to be downloaded, and the number of required computationally expensive signature verification operations. In addition to mathematical proofs, we have formalized the light client protocol in TLA+, and checked safety and liveness with the APALACHE model checker.