Aspect-Term Sentiment Analysis (ATSA) aims to predict sentiment polarity for specific aspect terms, a task complicated by conflicting sentiments and limited context in short texts. Existing graph-based approaches rely on predefined pairwise structures to capture different linguistic views. However, this leads to two key limitations: (1) their pairwise formulation often requires multiple graphs to improve expressive capacity, and (2) their reliance on predefined parsers or heuristic graph construction limits adaptability to sentence-specific sentiment composition. We propose HyperATSA, a dynamic hypergraph framework that overcomes these limitations through a single instance-specific hypergraph constructed directly from contextual token representations. Hyperedges are dynamically induced via hierarchical agglomerative clustering over token embeddings, where an acceleration-based cutoff identifies sentence-specific semantic groupings and enables adaptive hypergraph construction. Experiments on Lap14, Rest14, and MAMS demonstrate consistent improvements over strong graph-based baselines, suggesting that hypergraph-based relational modeling generalizes effectively to short-text sentiment composition.

Condensed summaries of newspaper articles cater to the modern need for easily digestible content amid shrinking attention spans. However, current summarization systems often produce extracts failing to capture the essence of original articles. Traditional evaluation metrics like ROUGE also provide limited insights into whether key information is preserved in the summaries. To address this, we propose a pipeline to generate high-quality summaries tailored for newspaper articles and evaluate them using a question-answering based metric. Our system segments input newspaper images, extracts text, and generates summaries. We also generate relevant questions from the original articles and use a question-answering model to assess how well the summaries can answer these queries to evaluate summary quality beyond just lexical overlap. Experiments on real-world data show the potential effectiveness of our approach in contrast to conventional metrics. Our framework holds promise for enabling reliable news summary generation and evaluation systems.