Nikolas Sargeant
Web3 security company Salus has revealed that its original research has been published in ‘Scientific Reports’ a sub-journal of the esteemed scientific journal, ‘Nature’.
In a press release shared with Cryptowisser, the team said its research paper examines the application of deep learning in smart contract vulnerability detection. The latest development marks a large step for Salus and web3 in general.
The research, titled “Deep learning-based solution for smart contract vulnerabilities detection” examines the potential for deep learning to identify code that is susceptible to attack through various exploits. The paper’s authors achieved an f1-score of 93.53% when using deep learning, showing their solution has the potential to outperform current methods.
A key problem with existing vulnerability checks is that they can produce false positives or negatives due to failure to accurately comprehend complex code logic. However, deep learning is not bound by predefined rules and can be continuously updated to be capable of detecting new attack vectors.
While commenting on this latest development, Salus co-founder Jiayi Li said:
“We’re grateful to ‘Scientific Reports’; a sub-journal of ‘Nature’ for publishing our research into the mitigation of smart contract vulnerabilities through deep learning. Throughout its 150-year history, Nature Publications have set a high bar for accepting peer-reviewed research into their journals. Its decision to publish the Salus team’s submission is a testament not only to the researchers’ methodology, but to the progression of web3. Salus continues to enhance the security advancement of Web3 through its ongoing technological research. As an advocate for the EVM Zero-Knowledge (ZK) application layer, Salus aids DAPPs on EVM in efficiently upgrading to ZK DAPPs.”
The research paper proposes a deep learning solution for identifying smart contract vulnerabilities called Lightning Cat. The solution leverages three deep learning models to maximize vulnerability identification.
The technology can also be used to pinpoint weaknesses in other types of code, giving it applications throughout the entire blockchain stack. Thanks to this, code dependability can be enhanced and the risk of exploitation significantly reduced, potentially preventing major financial loss.
Salus is a holistic web3 security company. With rich experience in traditional and blockchain security, it aims to solve some of the most complex security issues in the industry and make security accessible for all.
Meanwhile, Nature is a weekly international journal publishing the finest peer-reviewed research in all fields of science and technology on the basis of its originality, importance, interdisciplinary interest, timeliness, accessibility, elegance and surprising conclusions. Nature also provides rapid, authoritative, insightful and arresting news and interpretation of topical and coming trends affecting science, scientists and the wider public.
