We report our efforts in identifying a set of previous human evaluations in NLP that would be suitable for a coordinated study examining what makes human evaluations in NLP more/less reproducible. We present our results and findings, which include that just 13% of papers had (i) sufficiently low barriers to reproduction, and (ii) enough obtainable information, to be considered for reproduction, and that all but one of the experiments we selected for reproduction was discovered to have flaws that made the meaningfulness of conducting a reproduction questionable. As a result, we had to change our coordinated study design from a reproduce approach to a standardise-then-reproduce-twice approach. Our overall (negative) finding that the great majority of human evaluations in NLP is not repeatable and/or not reproducible and/or too flawed to justify reproduction, paints a dire picture, but presents an opportunity for a rethink about how to design and report human evaluations in NLP.

This paper presents a human evaluation reproduction study regarding the data-to-text generation task. The evaluation focuses in counting the supported and contradicting facts generated by a neural data-to-text model with a macro planning stage. The model is tested generating sport summaries for the ROTOWIRE dataset. We first describe the approach to reproduction that is agreed in the context of the ReproHum project. Then, we detail the entire configuration of the original human evaluation and the adaptations that had to be made to reproduce such an evaluation. Finally, we compare the reproduction results with those reported in the paper that was taken as reference.

The evaluation of Natural Language Generation (NLG) systems has recently aroused much interest in the research community, since it should address several challenging aspects, such as readability of the generated texts, adequacy to the user within a particular context and moment and linguistic quality-related issues (e.g., correctness, coherence, understandability), among others. In this paper, we propose a novel technique for evaluating NLG systems that is inspired on the triangular test used in the field of sensory analysis. This technique allows us to compare two texts generated by different subjects and to i) determine whether statistically significant differences are detected between them when evaluated by humans and ii) quantify to what extent the number of evaluators plays an important role in the sensitivity of the results. As a proof of concept, we apply this evaluation technique in a real use case in the field of meteorology, showing the advantages and disadvantages of our proposal.