Public awareness of Autism Spectrum Disorder (ASD) has grown in recent years, yet stigma surrounding this condition persists. Building on prior research showing increasingly positive portrayals of ASD, this study examines recent longitudinal trends in stigma and ASD, with a focus on Italian newspapers, and how these were affected by a key event such as the COVID-19 pandemic. We analyzed nearly 3,000 articles published between 2016 and 2025 using an innovative multi-layered Natural Language Processing (NLP) framework to capture multiple dimensions of stigma, including discriminatory language, emotional framings indicative of prejudices, stereotypes, and the thematic contexts in which ASD-related stigma appears. Overall, results indicate low levels of overt stigma and a gradual shift toward more positive portrayals, with only temporary disruptions during the pandemic. Some stereotypes remain, highlighting the need for ongoing attention to ASD representation in the media.

Although the affective dimension is a key aspect of metaphor, computational studies of figurative language have largely overlooked psycholinguistic variables such as valence. This study investigates whether computational models can reliably estimate the affective aspects of Italian and German metaphors and whether metaphor valence is compositionally derived. Outputs of BERT-based sentiment analysis and a valence-prompted LLM were compared with human ratings. Results show that the former exhibit limited alignment with human judgments, whereas higher agreement is achieved when the explicit concept of valence is prompted in a LLM. Both humans and models rely on the combined valence of the individual lemmas, suggesting a compositional contribution to metaphor valence.

The large availability of processable textual resources for Classical Latin has made it possible to study Latin literature through methods and tools that support distant reading. This paper describes a number of experiments carried out to test the possibility of investigating the thematic distribution of the Classical Latin corpus Opera Latina by means of topic modeling. For this purpose, we train, optimize and compare two neural models, Product-of-Experts LDA (ProdLDA) and Embedded Topic Model (ETM), opportunely revised to deal with the textual data from a Classical Latin corpus, to evaluate which one performs better both on the basis of topic diversity and topic coherence metrics, and from a human judgment point of view. Our results show that the topics extracted by neural models are coherent and interpretable and that they are significant from the perspective of a Latin scholar. The source code of the proposed model is available at https://github.com/MIND-Lab/LatinProdLDA.