Towards Finding Accounting Errors in Smart Contracts
Accepted to International Conference on Software Engineering 2024 (ICSE), 2024
In this paper, we motivate the detection of accounting errors in smart contracts by introducing an automated tool which utilized refinement types for type checking. The tool outperforms current state-of-the-art tools.
Authors: Brian Zhang (solo)
Demystifying Exploitable Bugs in Smart Contracts
Accepted to International Conference on Software Engineering (ICSE), 2023
Authors: Zhuo Zhang, Brian Zhang, Wen Xu, Zhiqiang Lin
A Systematic Study of Recent Smart Contract Vulnerabilities
Accepted to Crypto Economics Security Conference (CESC), 2022
In this paper, we study a large set of 502 unique exploitable bugs from 158 real-world contracts reported/exploited in the past year or so and aim to summarize their root causes and distributions.
Authors: Zhuo Zhang (co-first), Brian Zhang (co-first), Wen Xu, Zhiqiang Lin
Achieving Model Accuracy and Robustness by Adversarial Training with Batch Normalization
Accepted to International Conference on Tools with Artificial Intelligence (ICTAI), 2022
We propose a novel adversarial training method called norm shaping, in which a model always uses batch norms in both adversarial training and inference.
Authors: Brian Zhang, Shiqing Ma