The rapid growth of video content has increased the demand for automated news clip generation. However, producing concise, engaging, and editorially compliant clips remains challenging. Existing video summarization methods show promise, yet they often struggle to maintain semantic coherence, editorial quality, and robustness, limiting their real-world applicability. To address these limitations, we propose a robust video summarization framework for news clip generation. Our approach uniquely integrates: (1) a novel contrastive loss within Generative Adversarial Networks (GANs) to enhance stability and semantic coherence, (2) Bayesian uncertainty modeling to handle noise and improve generalization, and (3) a self-attention mechanism for context-aware keyframe selection. Additionally, we incorporate user-defined editorial rules in a post-processing module to align the summaries with editorial guidelines. Extensive experiments were conducted on two standard benchmarks (TVSum and SumMe) and a real-world newsroom dataset. The results show that the proposed method achieves state-of-the-art performance on both benchmarks, outperforming prior work. On the newsroom dataset, the framework generates clips that demonstrate strong alignment with human-edited content, achieving 87-91.5% fidelity. These results highlight the practical effectiveness of our method in real-world news clip production.

Quality journalism has become more important than ever due to the need for quality and trustworthy media outlets that can provide accurate information to the public and help to address and counterbalance the wide and rapid spread of disinformation. At the same time, quality journalism is under pressure due to loss of revenue and competition from alternative information providers. This vision paper discusses how recent advances in Artificial Intelligence (AI), and in Machine Learning (ML) in particular, can be harnessed to support efficient production of high-quality journalism. From a news consumer perspective, the key parameter here concerns the degree of trust that is engendered by quality news production. For this reason, the paper will discuss how AI techniques can be applied to all aspects of news, at all stages of its production cycle, to increase trust.

Abstract Journalism relies more and more on information and communication technology (ICT). ICT-basedjournalistic knowledge platformscontinuously harvest potentially news-relevant information from the Internet and make it useful for journalists. Because information about the same event is available from different sources and formats vary widely,knowledge graphsare emerging as a preferred technology for integrating, enriching, and preparing information for journalistic use. The paper explores how journalistic knowledge graphs can be augmented with support fornews angles, which can help journalists to detect newsworthy events and make them interesting for the intended audience. We argue that finding newsworthy angles on news-related information is an important example of a topical problem in information science: that of detecting interesting events and situations in big data sets and presenting those events and situations in interesting ways.

News angles are approaches to content presentation in journalism, where the journalist chooses which facts of an event to present. The News Angler project investigates how to computationally support the creation and selection of original news angles for a news event based on information from big data sources. At least two creative approaches are possible. One is to maintain a library of well-known news angles represented in a suitable modeling language, matching published reports on a current event to news angles in order to identify possible angles that have not yet been used. A second approach is not to represent news angles explicitly, instead matching the current event with previous events, and transferring angles from past to present reports by similarity and analogy. Both approaches are described and technologies needed to proceed in either direction are discussed.

No abstract available