Briefings·26 min
Jiashuo Liang · Guancheng Li
Introduces a novel gradient-based prompt-injection technique that can generate universal triggers to manipulate open-source Large Language Model (LLM) outputs.
open-sourceaillm+3
Guancheng Li
2 talks in the archive.
Filters+
Loading filters…
1–2 of 2 talks · loading…
Main Stage·40 min
Bohan Liu · Zheng Wang · GuanCheng Li
Demonstrates transforming publicly patched Chromium N-day vulnerabilities into remote code execution 0-days across popular Electron and CEF desktop applications.
chromiumelectronsupply-chain+2