EUVIMEDEUVIMED entdecken
Crossref

HalluClean: A Unified Framework to Combat Hallucinations in LLMs

Yaxin Zhao, Yu Zhang

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (42) · 36092-36100

DOI: 10.1609/aaai.v40i42.40926 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yaxin Zhao und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large language models (LLMs) have achieved impressive performance across a wide range of natural language processing tasks, yet they often produce hallucinated content that undermines factual reliability. To address this challenge, we introduce HalluClean, a lightweight and task-agnostic framework for detecting and correcting hallucinations in LLM-generated text. HalluClean adopts a reasoning-enhanced paradigm, explicitly decomposing the process into planning, execution, and revision stages to identify and refine unsupported claims. It employs minimal task-routing prompts to enable zero-shot generalization across diverse domains, without relying on external knowledge sources or supervised detectors. We conduct extensive evaluations on five representative tasks—question answering, dialogue, summarization, math word problems, and contradiction detection. Experimental results show that HalluClean significantly improves factual consistency and outperforms competitive baselines, demonstrating its potential to enhance the trustworthiness of LLM outputs in real-world applications.

Zitieren

APA

Zhao, Y., & Zhang, Y. (2026). HalluClean: A Unified Framework to Combat Hallucinations in LLMs. Proceedings of the AAAI Conference on Artificial Intelligence, 40(42), 36092-36100. https://doi.org/10.1609/aaai.v40i42.40926

Vancouver

Zhao Y, Zhang Y. HalluClean: A Unified Framework to Combat Hallucinations in LLMs. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(42):36092-36100. doi:10.1609/aaai.v40i42.40926

BibTeX
@article{Zhao2026,
  author = {Zhao, Yaxin and Zhang, Yu},
  title = {HalluClean: A Unified Framework to Combat Hallucinations in LLMs},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {42},
  pages = {36092--36100},
  doi = {10.1609/aaai.v40i42.40926}
}
Kostenlos starten →