Secret Source
"Secret Source" research broadly investigates methods for concealing and retrieving information, encompassing both data hiding within various media (text, speech) and protecting sensitive data from inference by AI models. Current efforts focus on developing algorithms and architectures, such as Bayesian neural networks and convolutional neural networks, to improve the security and efficiency of these processes, while also addressing the privacy vulnerabilities of large language models. This research is crucial for advancing secure communication and data storage, as well as mitigating privacy risks associated with increasingly prevalent AI systems.
Papers
October 20, 2024
October 27, 2023
March 16, 2023
January 31, 2023
January 30, 2023
October 29, 2022
October 28, 2022
July 25, 2022