Search papers, labs, and topics across Lattice.
School of Cyber Science and Technology
3
0
5
7
LLMs can now audit cross-chain smart contracts with expert-level precision, achieving 95% coverage of vulnerable projects by explicitly mirroring human reasoning processes.
Backdoor attacks in LLMs can be defused at inference time, without retraining or external data, by geometrically smoothing attention patterns to disrupt adversarial routing.
LLMs can now automatically validate smart contract vulnerabilities by generating and refining profit-driven exploits, significantly reducing the burden of manual verification.
