Nov 24, 2025, in Donga Science
The First Glimpse of Gemini's Rampaging Incarnation
We were interviewed by journalists from Donga Science, the longest-running science magazine of South Korea, for our work ReWiND.
Dec 4, 2025, in Robohub
Teaching Robot Policies Without New Demonstrations
In their paper ReWiND: Language-Guided Rewards Teach Robot Policies without New Demonstrations, which was presented at CoRL 2025, a team of researchers from USC introduce a framework for learning robot manipulation tasks solely from language instructions without per-task demonstrations.
Oct 1, 2025, in USC Viterbi News
Robots, Meet Your New Teacher: Yourselves
A USC Viterbi graduate student team co-developed a new robotic system that adapts and learns every second.
Dec 11, 2023, in USC Viterbi News
Once is enough - Helping robots learn quickly in new environments
A new algorithm developed by USC computer science researchers shows that robots can, in computer simulations, learn tasks after a single demonstration
Oct 24, 2019, in TechXplore
A user-friendly approach for active reward learning in robots
In recent years, researchers have been trying to develop methods that enable robots to learn new skills. One option is for a robot to learn these new skills from humans, asking questions whenever it is unsure about how to behave, and learning from the human user's responses.
Jun 24, 2019, in Stanford News
Stanford researchers teach robots what humans want
Researchers are developing better, faster ways of providing human guidance to autonomous robots.
