Perspective: LLM Evaluation Beyond BLEU/ROUGE
Abstract
We propose evaluation metrics for large language models based on factuality, consistency, and latency-cost tradeoffs.
Cite this article
Gupta, S. & Santos, S. (2022). Perspective: LLM Evaluation Beyond BLEU/ROUGE. Research Explorations in Global Knowledge & Technology (REGKT), 1 (10). Retrieved from https://regkt.com/article.php?id=703&slug=perspective-llm-evaluation-beyond-bleu-rouge-v2