Emerging1d agoNew Research Method Improves AI Model Transparency and Safety Through Self-Consistency TrainingTechnologyTech Breakthrough66%