Currently, mainstream AI alignment methods such as Reinforcement Learning from Human Feedback (RLHF) and Direct Preference Optimization (DPO) rely on high-quality human preference feedback data.

Researchers developed a faster, more stable way to simulate the swirling electric fields inside industrial plasmas -- the kind used to make microchips and coat materials. The improved method could ...

A team of computer scientists at UC Riverside has developed a method to erase private and copyrighted data from artificial ...

UC Riverside researchers have created a certified unlearning method that removes sensitive or copyrighted data from AI models ...