We investigate how language models handle the proviso problem, an unresolved issue in pragmatics where presuppositions in conditional sentences diverge between theoretical and human interpretations. We reformulate this phenomenon as a Natural Language Inference task and introduce a diagnostic dataset designed to probe presupposition projection in conditionals. We evaluate RoBERTa, DeBERTa, LLaMA, and Gemma using explainability analyses. The results show that models broadly align with human judgments but rely on shallow pattern matching rather than semantic or pragmatic reasoning. Our work provides the first computational evaluation framework for the proviso problem and highlights the need for diagnostic, multi-method approaches to assess pragmatic competence and context-dependent meaning in language models.

The increasing suicide rates amongst youth and its high correlation with suicidal ideation expression on social media warrants a deeper investigation into models for the detection of suicidal intent in text such as tweets to enable prevention. However, the complexity of the natural language constructs makes this task very challenging. Deep Learning architectures such as LSTMs, CNNs, and RNNs show promise in sentence level classification problems. This work investigates the ability of deep learning architectures to build an accurate and robust model for suicidal ideation detection and compares their performance with standard baselines in text classification problems. The experimental results reveal the merit in C-LSTM based models as compared to other deep learning and machine learning based classification models for suicidal ideation detection.

Technological advancements in the World Wide Web and social networks in particular coupled with an increase in social media usage has led to a positive correlation between the exhibition of Suicidal ideation on websites such as Twitter and cases of suicide. This paper proposes a novel supervised approach for detecting suicidal ideation in content on Twitter. A set of features is proposed for training both linear and ensemble classifiers over a dataset of manually annotated tweets. The performance of the proposed methodology is compared against four baselines that utilize varying approaches to validate its utility. The results are finally summarized by reflecting on the effect of the inclusion of the proposed features one by one for suicidal ideation detection.