Timezone: »

Automatic Construction of Evaluation Suites for Natural Language Generation Datasets
Simon Mille · Kaustubh Dhole · Saad Mahamood · Laura Perez-Beltrachini · Varun Prashant Gangal · Mihir Kale · Emiel van Miltenburg · Sebastian Gehrmann

Machine learning approaches applied to NLP are often evaluated by summarizing their performance in a single number, for example accuracy. Since most test sets are constructed as an i.i.d. sample from the overall data, this approach overly simplifies the complexity of language and encourages overfitting to the head of the data distribution. As such, rare language phenomena or text about underrepresented groups are not equally included in the evaluation. To encourage more in-depth model analyses, researchers have proposed the use of multiple test sets, also called challenge sets, that assess specific capabilities of a model. In this paper, we develop a framework based on this idea which is able to generate controlled perturbations and identify subsets in text-to-scalar, text-to-text, or data-to-text settings. By applying this framework to the GEM generation benchmark, we propose an evaluation suite made of 80 challenge sets, demonstrate the kinds of analyses that it enables and shed light onto the limits of current generation models.

Author Information

Simon Mille (Universitat Pompeu Fabra)
Kaustubh Dhole (BITS Pilani)
Saad Mahamood
Laura Perez-Beltrachini (University of Edinburgh)
Varun Prashant Gangal (Carnegie Mellon University)
Mihir Kale (Carnegie Mellon University)
Emiel van Miltenburg (Tilburg University)
Sebastian Gehrmann (Google Research)

More from the Same Authors