These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.
Google BLEU (GLEU)
The BLEU score has some undesirable properties when used for single sentences, as it was designed to be a corpus measure. The Google BLEU score, also known as GLEU score, is designed to limit these undesirable properties when used for single sentences.
To calculate this score, all sub-sequences of 1, 2, 3 or 4 tokens in output and target sequence (n-grams) are recorded. The precision and recall, described below, are then computed.
- precision: the ratio of the number of matching n-grams to the number of total n-grams in the generated output sequence
- recall: the ratio of the number of matching n-grams to the number of total n-grams in the target (ground truth) sequence
The minimum value of precision and recall is then returned as the score.
About the metric
You can click on the links to see the associated metrics
Objective(s):
Purpose(s):
Lifecycle stage(s):
Target users:
Risk management stage(s):
