Learned language-conditioned robot policies often struggle to effectively adapt to new real-world tasks even when pre-trained across a diverse set of instructions. We propose a novel approach for few-shot adaptation to unseen tasks that exploits the semantic understanding of task decomposition provided by vision-language models (VLMs). Our method, Policy Adaptation via Language Optimization (PALO), combines a handful of demonstrations of a task with proposed language decompositions sampled from a VLM to quickly enable rapid nonparametric adaptation, avoiding the need for a larger fine-tuning dataset. We evaluate PALO on extensive real-world experiments consisting of challenging unseen, long-horizon robot manipulation tasks. We find that PALO is able of consistently complete long-horizon, multi-tier tasks in the real world, outperforming state of the art pre-trained generalist policies, and methods that have access to the same demonstrations.

Can artificial intelligence replace the human brain?Will it? "Humans were are not built to spend more than two hours looking at a screen or scrolling through excel sheets. Humans are best at being human. Artificial Intelligence will do the rest." Telling words from Jim Stolze, Co-founder of aigency -- an Amsterdam-based company that recruits AI and humans for work.

"Humans were are not built to spend more than two hours looking at a screen or scrolling through excel sheets. Humans are best at being human. Artificial Intelligence will do the rest." Kind of an employment company run by three humans overseeing 59 robots (actually computers working on algorithms created at the University of Amsterdam to solve problems). Stolze was addressing reporters in StartUp Village at the Amsterdam Science Park on the sidelines of the first World Summit AI in Amsterdam October 11-12.

Can artificial intelligence replace the human brain?Will it? "Humans were are not built to spend more than two hours looking at a screen or scrolling through excel sheets. Humans are best at being human. Artificial Intelligence will do the rest." Kind of an employment company run by three humans overseeing 59 robots (actually computers working on algorithms created at the University of Amsterdam to solve problems).