ROUGE-L

What is ROUGE-L?

Evaluation metric measuring the longest common subsequence between machine-generated text and reference summaries to assess content overlap

In other languages

한국어ROUGE-L

모델이 생성한 텍스트와 정답 텍스트 간의 최장 공통 부분 수열을 측정하여 요약 및 번역 성능을 평가하는 지표

日本語ROUGE-L

機械翻訳や要約タスクにおいて、生成文と参照文の最長共通部分列を用いて精度を測定する評価指標。

Related Terms

• ROUGEText evaluation metric family that compares generated summaries with reference summaries using overlapping units
• BLEUTranslation evaluation metric that compares machine output with reference translations using n-gram overlap
• AA-LCRArtificial Analysis evaluation focused on language comprehension and reasoning performance
• Character error rateA speech-recognition metric that measures the proportion of characters recognized incorrectly compared with a reference transcript.
• LLM judgeA language model tasked with evaluating the quality, accuracy, or adherence of another model's output to a given rubric