Adapting driving behavior to new environments, customs, and laws is a long-standing challenge in autonomous driving. LLaDA (Large Language Driving Assistant) is an #LLM network that makes it easier to navigate in unfamiliar places by providing real-time guidance on regional traffic rules in different languages, for both human drivers and #autonomousvehicles. LLaDA will be powered by NVIDIA DRIVE Thor, which harnesses the new #generativeAI capabilities of NVIDIA’s Blackwell GPU architecture.
00:00:12 – Introducing LLaDA
00:00:36 – Multimodal neural interfaces integrate diverse types of data
00:00:53 – LLaDA can rapidly adapt to local traffic rules and customs
00:01:30 – The NVIDIA Riva speech SDK can process different languages
00:01:58 – LLaDA can also be applied to AV motion planning
00:02:28 – Accelerated by NVIDIA DRIVE Thor, built on the NVIDIA Blackwell architecture
00:03:02 – Visit our GitHub page and check out the LLaDA paper at CVPR 2024
Project page: https://boyiliee.github.io/llada/
Paper: https://arxiv.org/abs/2402.05932
Watch the full series here: https://nvda.ws/3LsSgnH
Learn more about DRIVE Labs: https://nvda.ws/36r5c6t
Follow us on social:
Twitter: https://nvda.ws/3LRdkSs
LinkedIn: https://nvda.ws/3wI4kue
#NVIDIADRIVE
Please Support our Sponsors here ---