About this event

We will take a dive into the field of evaluation of generated text where Philipp Koch will present his paper “Pre-trained language models evaluating themselves - A comparative study” (https://aclanthology.org/2022.insights-1.25/), which was recently accepted to the Workshop on Insights from Negative Results in NLP @ACL'22.

Evaluation of generated text remains a significant issue. Recently introduced model-based metrics have shown promising results compared to n-gram-based metrics like BLEU, yet they still suffer severe drawbacks.

Speaker

Philipp Koch is one of the founders of Munich NLP. He is currently studying M.Sc. Statistics and Data Science at LMU.