Unfortunately, I see many students (and indeed other people) make some basic mistakes when evaluating machine learning, for classifiers as well as NLG.
I was recently asked by someone if it was possible to easily determine whether an NLP system was good enough for a specific use case. Currently this is very hard. Making it easy could be a “grand challenge” for evaluation!
In both NLG and MT contexts, deep learning approaches can result in texts which are fluent and readable but also incorrect and misleading. This is problematical if accuracy is more important than readability, as is the case in most NLG contexts.
Many neural NLG systems “hallucinate” non-existent or incorrect content. This is a major problem, since such hallucination is unacceptable in many (most?) NLG use cases. Also BLEU and related metrics do not detect hallucination well, so researchers who rely on such metrics may be misled about the quality of their system.
Unfortunately I suspect many researchers make their results looks better by using poor baselines. I give some thoughts on this, based on a recent discussion with a PhD student.
Some musings on principled and theoretically sound techniques for automatically evaluating NLG systems.
My advice on how to perform a high-quality validation study, which assesses whether a metric (such as BLEU) correlates well with human evaluations.
BLEU works much better for MT systems and NLG systems. In this blog I present some speculations as to why this is the case.
My structured survey of BLEU suggests that BLEU-human correlations are worse in German than in many other languages. But there are many caveats, so we need to be cautious in interpreting this result.
The correlation between BLEU and human evaluations of MT systems seems to be increasing over time. Since BLEU has not changed, how is this possible, and what does it mean?