1 code implementation • 5 Apr 2024 • Gawon Choi, Hyemin Ahn
This leads us to pose a question: If small LMs can be trained to reason in chains within a single domain, would even small LMs be good task planners for the robots?