23 citations · 48 across the 17 of their papers we have counts for
3 papers · 1 filter
BMX: Boosting Natural Language Generation Metrics with Explainability
Christoph Leiter, Hoa Nguyen, Steffen Eger
State-of-the-art natural language generation evaluation metrics are based on black-box language models. Hence, recent works consider their explainability with the goals of better u…
EffEval: A Comprehensive Evaluation of Efficiency for MT Evaluation Metrics
Daniil Larionov, Jens Grünwald, Christoph Leiter +1
Efficiency is a key property to foster inclusiveness and reduce environmental costs, especially in an era of LLMs. In this work, we provide a comprehensive evaluation of efficiency…
Towards Explainable Evaluation Metrics for Natural Language Generation
Christoph Leiter, Piyawat Lertvittayakumjorn, Marina Fomicheva +3
Unlike classical lexical overlap metrics such as BLEU, most current evaluation metrics (such as BERTScore or MoverScore) are based on black-box language models such as BERT or XLM-…