Culturally Grounded Physical Commonsense Reasoning in Italian and English: A Submission to the MRL 2025 Shared Task

TL;DR

This paper introduces FormaMentis, a culturally grounded benchmark for physical commonsense reasoning in Italian and English, created through expert annotations that incorporate local customs and norms, advancing multilingual AI understanding.

Contribution

It presents a novel culturally grounded dataset for physical commonsense reasoning in Italian and English, filling a gap in multilingual and culturally aware AI benchmarks.

Findings

FormaMentis enables evaluation of physical reasoning across cultures.

Expert annotations ensure cultural relevance and accuracy.

The dataset supports multilingual and cross-cultural AI research.

Abstract

This paper presents our submission to the MRL 2025 Shared Task on Multilingual Physical Reasoning Datasets. The objective of the shared task is to create manually-annotated evaluation data in the physical commonsense reasoning domain, for languages other than English, following a format similar to PIQA. Our contribution, FormaMentis, is a novel benchmark for physical commonsense reasoning that is grounded in Italian language and culture. The data samples in FormaMentis are created by expert annotators who are native Italian speakers and are familiar with local customs and norms. The samples are additionally translated into English, while preserving the cultural elements unique to the Italian context.