EUVIMEDEUVIMED entdecken
Crossref

Evaluating the Factuality of Large Language Models Using Multiple Plug-and-Play Fact Sources

Zhaoheng Huang, Yutao Zhu, Jirong Wen, Zhicheng Dou

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (48) · 41607-41609

DOI: 10.1609/aaai.v40i48.42355 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Zhaoheng Huang und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large language models (LLMs) often produce factually inaccurate content, or hallucinations, which undermines their reliability. Existing factuality evaluation systems usually rely on a single predefined fact source, making them task-specific and hard to extend. We present UFO, a unified framework for factuality evaluation that supports multiple plug-and-play fact sources. UFO integrates human-written evidence, web search results, and LLM knowledge within a single evaluation pipeline, and allows users to flexibly select, reorder, and even define customized sources. The system is accessible through both a Python interface and a web-based demo, offering interactive claim-level verification and visualization. Experiments show that UFO system achieves moderate consistency with human annotations. Overall, UFO serves as a transparent and extensible platform for benchmarking fact sources, comparing LLMs, and enabling real-world fact-checking applications across diverse domains.

Zitieren

APA

Huang, Z., Zhu, Y., Wen, J., & Dou, Z. (2026). Evaluating the Factuality of Large Language Models Using Multiple Plug-and-Play Fact Sources. Proceedings of the AAAI Conference on Artificial Intelligence, 40(48), 41607-41609. https://doi.org/10.1609/aaai.v40i48.42355

Vancouver

Huang Z, Zhu Y, Wen J, Dou Z. Evaluating the Factuality of Large Language Models Using Multiple Plug-and-Play Fact Sources. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(48):41607-41609. doi:10.1609/aaai.v40i48.42355

BibTeX
@article{Huang2026,
  author = {Huang, Zhaoheng and Zhu, Yutao and Wen, Jirong and Dou, Zhicheng},
  title = {Evaluating the Factuality of Large Language Models Using Multiple Plug-and-Play Fact Sources},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {48},
  pages = {41607--41609},
  doi = {10.1609/aaai.v40i48.42355}
}
Kostenlos starten →