Crypto & Blockchain · Wed, Nov 11, 2026

Bitcoin, Ethereum, stablecoins, DeFi, smart contracts, scaling and consensus.

RSS feed
arXiv cs.CL

AraCrypto-ML: A Multi-Label Corpus of Communicative Functions in Arabic Facebook Cryptocurrency Discourse

In briefAraCrypto-ML: A Multi-Label Corpus of Arabic Facebook Cryptocurrency Posts by Communicative Function

SummaryThis paper introduces AraCrypto-ML, a corpus of 10,000 multi-label annotated Arabic Facebook posts about cryptocurrency from 2014 to 2024, drawn from 2,898 public pages. Posts are labelled into six functional categories using the Fanar-2-27B model under a prompt-only protocol. The authors validate 1,000 posts against adjudicated gold labels and apply Rogan-Gladen calibration to correct labelling bias. Reaction patterns vary markedly across functions, with Love dominating Educational and Promotional posts and Haha most common on Skeptical-Warning posts.

Mabrouka Bessghaier, Mohamed Rashad Rouine, Md. Rafiul Biswas et al.

arXiv cs.DC · cs.CR · cs.DB

CuEVM: Scaling Smart Contract Execution on GPUs

In briefCuEVM: Scaling Smart Contract Execution on GPUs

SummaryEthereum smart contracts process transactions worth billions of dollars daily, but the Ethereum Virtual Machine executes them sequentially, creating a performance bottleneck. This paper presents CuEVM, a GPU-native and semantically complete EVM execution engine that treats transaction execution as a massively parallel data processing problem. It passes 100% of the Ethereum VMTests and 96.2% of the full execution-test suite, and executes over 8 million ERC-20 transfers per second on a single commodity GPU. The authors also built CuEVM-fuzz, a smart contract fuzzer on top of CuEVM, which achieves over a million end-to-end executions per second.

Nhut-Minh Ho, Tien Tuan Anh Dinh, Quang Trung Ta et al.