Recognizing fallacies in everyday argumentation plays a key role in developing individuals' critical thinking skills, ultimately contributing to mitigate faulty and harmful argumentation at its root.

FadeIT is the first shared task on fallacy detection for Italian in public discourse about migration, climate change, and public health over a 4-year time period. It is part of the EVALITA 2026 campaign, whose workshop will be held in Bari, Apulia (Italy) on February 26-27, 2026.

FadeIT is based on Faina (Ramponi et al., 2025), a dataset for fallacy detection in Italian social media posts that includes disaggregated span-level annotations with overlaps for 20 fallacy types. Participant teams can leverage disaggregated annotations or aggregate them to develop systems to detect fallacies in subtasks of increasing complexity: at the post level (coarse-grained fallacy detection) and/or at the span level (fine-grained fallacy detection). Systems will be evaluated against two equally valid gold standards to account for genuine disagreement, in line with work advocating the importance of valuing human label variation in subjective tasks (Plank, 2022).

Learn more on the task (w/ associated subtasks), the data (w/ examples), and how to participate!