@misc{13730,
  abstract     = {{This paper introduces an LLM-mediated AI Advisor that contextualizes and synthesizes heterogeneous explainable AI (XAI) outputs to support fast and calibrated misinformation judgments in time-sensitive social media settings. We define LLM-mediated XAI as a process in which a large language model aggregates, prioritizes, and translates heterogeneous XAI outputs into a context-sensitive explanation tailored to the user’s decision situation. Semantic features, XAI modules and LLM-based summarization and synthesis enable the generation of explanations that are adapted in three ways: compressed for time-efficient decisions, translated into non-technical language, and progressively expandable for deeper inspection. Through a mixed-methods user study, including a quantitative study and a qualitative study, we analyze how users interpret, challenge and strategically rely on LLM-mediated explanations during real-world misinformation assessment tasks. The findings indicate that the approach reduces time-to-decision and supports critical inspection without inducing over-reliance. Progressive disclosure and different techniques to present information favored different user needs while conversational functionality was rarely used due to unclear benefits and fear of confusion.}},
  author       = {{Grimm, Valentin and Rubart, Jessica and Herder, Eelco and Röcker, Carsten}},
  booktitle    = {{WebSci Companion '26: Companion Publication of the 2026 18th ACM Web Science Conference}},
  editor       = {{Balke, Wolf-Tilo and Plötzky, Florian and Spaniol, Marc and Herder, Eelco and Manikonda, Lydia and Liu, Haiming and Ibáñez, Luis-Daniel and Rezapour, Rezvaneh}},
  isbn         = {{979-8-4007-2492-3}},
  keywords     = {{Large Language Model Mediation, Explainable AI, Decision Co- Pilot Systems, Misinformation Detection}},
  location     = {{Braunschweig}},
  pages        = {{110--116}},
  publisher    = {{ACM}},
  title        = {{{LLM-Mediated XAI Explanations: An AI Advisor for Fast and Calibrated Judgments on Potential Misinformation}}},
  doi          = {{https://doi.org/10.1145/3795513.3810452}},
  year         = {{2026}},
}

@misc{13529,
  abstract     = {{The proliferation of misinformation is one of the most pressing challenges in today’s digital landscape, due to its far-reaching implications for public health, economic stability, trust in governmental institutions, and societal cohesion. Despite efforts to regulate online platforms and limit the spread of misinformation, many individuals are left behind because of their low digital literacy, level of education, and other contributing factors. In this context, we explore the use of Large Language Models (LLMs) to identify misinformation and we evaluate the capabilities of GPT-4.1-mini, as a representative example of these models. We then discuss how LLMs can help empower users to critically create and share information, thereby fostering more resilient online communities. We also present a set of possible interaction patterns for content creation and moderation.}},
  author       = {{Franco, Mirko and Grimm, Valentin and Herder, Eelco}},
  booktitle    = {{Proceedings of the 2025 International Conference on Information Technology for Social Good}},
  editor       = {{Marquez-Barja, Johann and Bujari, Armir and Slamnik-Kriještorac, Nina and Sabbioni, Andrea}},
  isbn         = {{979-8-4007-2089-5}},
  keywords     = {{misinformation, fake news, large language models, online social networks}},
  location     = {{Antwerp, Belgium}},
  pages        = {{244 -- 252}},
  publisher    = {{ACM}},
  title        = {{{Preventing Accidental Sharing of Misinformation Using Large Language Models}}},
  doi          = {{10.1145/3748699.3749798}},
  year         = {{2025}},
}