参考文献:[1] Do As I Can, Not As I Say: Grounding Language in Robotic Affordances, https://arxiv.org/pdf/2204.01691.pdf[2] Towards Helpful Robots: Grounding Language in Robotic Affordances,https://ai.googleblog.com/2022/08/towards-helpful-robots-grounding.html[3] Grounding language in robotic affordances, https://www.youtube.com/watch?v=E2R1D8RzOlM[4] Language Models are Few-Shot Learners, https://arxiv.org/abs/2005.14165[5] Introducing FLAN: More generalizable Language Models with Instruction Fine-Tuning, https://ai.googleblog.com/2021/10/introducing-flan-more-generalizable.html[6] Pathways Language Model (PaLM): Scaling to 540 Billion Parameters for Breakthrough Performance, https://ai.googleblog.com/2022/04/pathways-language-model-palm-scaling-to.html[7] PaLM-SayCan is also interpretable, https://github.com/say-can/say-can.github.io/blob/main/img/demo_sequence_compressed.mp4?raw=true