Is Model Attention Aligned with Human Attention? An Empirical Study on Large Language Models for Code Generation

Open in new window