Performance metrics (Pass@k) for one-shot, few-shot, and zero-shot LLM generations evaluated on React-based dashboard components.
Each entry reports the probability that at least one of the top-k generated outputs compiles successfully.
Number of samples: 81
Tags: evaluation, pass@k, react, llm, one-shot, few-shot, zero-shot
License: MIT
