We present two comprehensive benchmarks to evaluate the performance of language models in coding assistance tasks, covering code writing, debugging, code review, and conceptual understanding. Our main ...

Abstract: To execute a variety of collaborative tasks, the cooperation for unmanned aerial vehicles (UAVs) with complicated interactions under dynamic environments is a challenging and critical issue.