Post

SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Safe Reinforcement Learning

๐Ÿ“Œ ์—ฐ๊ตฌ ๊ฐœ์š”

Confidence Calibration์€ ๋ชจ๋ธ์ด ์˜ˆ์ธกํ•œ ํ™•๋ฅ ์ด ์‹ค์ œ ์ •๋‹ต์ผ ๊ฐ€๋Šฅ์„ฑ๊ณผ ์–ผ๋งˆ๋‚˜ ์ผ์น˜ํ•˜๋Š”์ง€๋ฅผ ๋‚˜ํƒ€๋‚ด๋Š” ๊ฐœ๋…์ด๋‹ค.

This post is licensed under CC BY 4.0 by the author.