Narrative similarity is a challenging problem that requires reasoning over three aspects of narratives, including (1) the abstract theme; (2) the course of action and (3) the outcomes of narratives. We present UTD.HLTRISIM.NARRATIVES, our method developed for SemEval 2026 Task 4 (Narrative Story Similarity), which combines contrastive reasoning prompting with careful selection of few-shot examples to guide a Large Language Model(LLM) toward decisions of narrative comparative similarity. A curriculum learning framework orders examples of narrative triplets presented to the LLM by using a score that quantifies the impact of common narratives aspects with information discerned from several distractors of narrative similarity between pairs ofnarratives 1.

Though Large Language Models (LLMs) have been serving global users through a wide range of services, concerns remain regarding their cultural bias and misalignment with people of underrepresented communities. Increasing use of LLMs presents significant implications, as they have the potential to influence people’s original values toward a certain cultural perspective. Cultural alignment of LLMs with culture-specific knowledge offers a suitable solution to this concern. In our participation in the Semeval-2026 Task 7 we considered a prompt engineering-based cultural alignment strategy to address the cultural knowledge gap in LLMs. Our approach achieved promising 86.34% accuracy for Japanese culture-relevant multiple-choice questions from the BLEND benchmark.

Misogyny, which is widespread on social media, can be identified not only by recognizing its many forms but also by discovering how misogyny is framed. This paper considers the automatic discovery of misogyny problems and their frames through the Dis-MP&F method, which enables the generation of a data-driven, rich Taxonomy of Misogyny (ToM), offering new insights in the complexity of expressions of misogyny. Furthermore, the Dis-MP&F method, informed by the ToM, is capable of producing very promising results on a misogyny benchmark dataset.

Women are frequently targeted online with hate speech and misogyny using tweets, memes, and other forms of communication. This paper describes our system for Task 5 of SemEval-2022: Multimedia Automatic Misogyny Identification (MAMI). We participated in both the sub-tasks, where we used transformer-based architecture to combine features of images and text. We explore models with multi-modal pre-training (VisualBERT) and text-based pre-training (MMBT) while drawing comparative results. We also show how additional training with task-related external data can improve the model performance. We achieved sizable improvements over baseline models and the official evaluation ranked our system 3^rd out of 83 teams on the binary classification task (Sub-task A) with an F1 score of 0.761, and 7^th out of 48 teams on the multi-label classification task (Sub-task B) with an F1 score of 0.705.

Patronizing behavior is a subtle form of bullying and when directed towards vulnerable communities, it can arise inequalities. This paper describes our system for Task 4 of SemEval-2022: Patronizing and Condescending Language Detection (PCL). We participated in both the sub-tasks and conducted extensive experiments to analyze the effects of data augmentation and loss functions used, to tackle the problem of class imbalance. We explore whether large transformer-based models can capture the intricacies associated with PCL detection. Our solution consists of an ensemble of the RoBERTa model which is further trained on external data and other language models such as XLNeT, Ernie-2.0, and BERT. We also present the results of several problem transformation techniques such as Classifier Chains, Label Powerset, and Binary relevance for multi-label classification.