International Workshop on Blockchain and Web3 for Education, Research, and Science

Xiaofeng Zheng; Yinfeng Cao

This paper proposes a security and privacy protection framework for traffic large models that integrates consortium blockchain, federated learning, adaptive differential privacy, and zero-knowledge proof. The framework records model versions and gradient data for trusted lifecycle traceability while keeping sensitive traffic data within local domains. Experiments report model tampering detection within 200 ms, a 68% reduction in membership inference attack success rate, and a 98.7% gradient poisoning interception rate.

• Traffic large models
• Blockchain
• Federated learning
• Privacy protection
• Zero-knowledge proof

Huiyu Bao; Changze Zheng; Yuwei Ji; Kevin Cao

This paper studies collaborative load forecasting across distribution network station areas where data are distributed across multiple control units. It combines a DDC-LSTNet-skip forecasting network with blockchain-based trust management, off-chain storage, on-chain notarization, and smart-contract access control. The proposed method improves forecasting accuracy while strengthening secure data interaction, result traceability, and operational reliability.

• Load forecasting
• Blockchain
• Spatio-temporal modeling
• Smart contracts
• Traceability

Xingxi Ding; Hongyu Liu

This survey organizes blockchain-based cross-border applications across payments, trade finance, identity exchange, and tokenized settlement. It presents a taxonomy covering application objectives, governance structures, settlement models, interoperability mechanisms, privacy designs, and regulatory constraints. The paper argues that practical value depends on off-chain enforcement, data standardization, jurisdictional compatibility, and operational governance as much as the ledger itself.

• Survey
• Blockchain
• Cross-border applications
• Governance
• Interoperability