Enhancing Safety

Enhancing safety across various autonomous systems, particularly in autonomous vehicles and large language models (LLMs), is a central research focus. Current efforts concentrate on improving model robustness through techniques like diffusion models for generating realistic safety-critical scenarios, control barrier functions combined with neural radiance fields for visual feedback control, and parameter-efficient fine-tuning methods (e.g., LoRA) for LLMs that mitigate safety risks during training. These advancements aim to create more reliable and trustworthy systems, impacting fields ranging from transportation to AI safety and ultimately contributing to safer and more efficient technologies.