The F1 Score at Top-K is an important evaluation metric for keyword extraction tasks in the field of natural language processing. F1@K calculates the harmonic mean of precision and recall over the Top-K keywords, comprehensively evaluating ranking and extraction quality. It is the most commonly used comprehensive evaluation standard for keyword extraction, balancing the impact of missed extractions and false positives. It is typically reported as a percentage or a score between 0 and 1. This metric is widely used in relevant benchmarks and papers, effectively reflecting model performance in this dimension. Higher values indicate better performance.