Many researchers have reached the conclusion that AI models should be trained to be aware of the possibility of variation and disagreement in human judgments, and evaluated as per their ability to recognize such variation. The LEWIDI series of shared tasks on Learning With Disagreements was established to promote this approach to training and evaluating AI models, by making suitable datasets more accessible and by developing evaluation methods. The third edition of the task builds on this goal by extending the LEWIDI benchmark to four datasets spanning paraphrase identification, irony detection, sarcasm detection, and natural language inference, with labeling schemes that include not only categorical judgments as in previous editions, but ordinal judgments as well. Another novelty is that we adopt two complementary paradigms to evaluate disagreement-aware systems: the soft-label approach, in which models predict population-level distributions of judgments, and the perspectivist approach, in which models predict the interpretations of individual annotators. Crucially, we moved beyond standard metrics such as cross-entropy, and tested new evaluation metrics for the two paradigms. The task attracted diverse participation, and the results provide insights into the strengths and limitations of methods to modeling variation. Together, these contributions strengthen LEWIDI as a framework and provide new resources, benchmarks, and findings to support the development of disagreement-aware technologies.

There are two competing approaches for modelling annotator disagreement: distributional soft-labelling approaches (which aim to capture the level of disagreement) or modelling perspectives of individual annotators or groups thereof. We adapt a multi-task architecture -- which has previously shown success in modelling perspectives -- to evaluate its performance on the SEMEVAL Task 11. We do so by combining both approaches, i.e. predicting individual annotator perspectives as an interim step towards predicting annotator disagreement. Despite its previous success, we found that a multi-task approach performed poorly on datasets which contained distinct annotator opinions, suggesting that this approach may not always be suitable when modelling perspectives. Furthermore, our results explain that while strongly perspectivist approaches might not achieve state-of-the-art performance according to evaluation metrics used by distributional approaches, our approach allows for a more nuanced understanding of individual perspectives present in the data. We argue that perspectivist approaches are preferable because they enable decision makers to amplify minority views, and that it is important to re-evaluate metrics to reflect this goal.

Most Artificial Intelligence applications are based on supervised machine learning (ML), which ultimately grounds on manually annotated data. The annotation process is often performed in terms of a majority vote and this has been proved to be often problematic, as highlighted by recent studies on the evaluation of ML models. In this article we describe and advocate for a different paradigm, which we call data perspectivism, which moves away from traditional gold standard datasets, towards the adoption of methods that integrate the opinions and perspectives of the human subjects involved in the knowledge representation step of ML processes. Drawing on previous works which inspired our proposal we describe the potential of our proposal for not only the more subjective tasks (e.g. those related to human language) but also to tasks commonly understood as objective (e.g. medical decision making), and present the main advantages of adopting a perspectivist stance in ML, as well as possible disadvantages, and various ways in which such a stance can be implemented in practice. Finally, we share a set of recommendations and outline a research agenda to advance the perspectivist stance in ML.