The field of social media analysis and automated systems is rapidly evolving, with a significant focus on leveraging large language models to improve the accuracy and efficiency of various tasks. Researchers are exploring innovative approaches to analyze and understand online discourse, including the development of frameworks for automated ticket escalation, truthfulness stance mapping, and emotion alignment. These studies demonstrate the potential of large language models to enhance our understanding of social media dynamics and improve the management of online platforms. Notably, the integration of multimodal large language models is showing promise in predicting the perceived credibility of visual content and explaining human judgment. Furthermore, the application of collections demography and citizen science approaches is providing new insights into the preservation and management of cultural heritage sites. Some noteworthy papers in this area include:
- TickIt, which introduces an innovative online ticket escalation framework powered by large language models,
- LLMTaxo, which leverages large language models for the automated construction of taxonomy of factual claims from social media,
- Integrating Emotion Distribution Networks and Textual Message Analysis for X User Emotional State Classification, which employs hybrid methodologies to improve emotion classification accuracy.