EUVIMEDEUVIMED entdecken
Crossref

AuditAgent: LLM Agent for Risks Auditing in Recommender Systems

Du Su, Zhenxing Chen, Shilong Zhao, Yuanhao Liu, Fei Sun, Qi Cao, Huawei Shen

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (48) · 41688-41690

DOI: 10.1609/aaai.v40i48.42382 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Du Su und 6 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Auditing recommendation systems has attracted growing attention due to increasing concerns over filter bubbles, unfairness, and data misuse. A common approach is sock-puppet auditing, where autonomous agents interact with platforms to reveal risks. However, existing approaches rely on hard-coded agents, lacking adaptability to dynamic GUI layouts and generating behaviors far from those of real users, limiting the comprehensiveness and representativeness of assessment. To address these issues, we introduce AuditAgent, an LLM-powered GUI-agent framework for risk auditing. AuditAgent simulates realistic user preferences and performs adaptive, human-like interactions on recommendation platforms. This design enables more thorough and faithful auditing, providing comprehensive assessments across multiple risk dimensions, including filter bubbles, unfairness, and data misuse.

Zitieren

APA

Su, D., Chen, Z., Zhao, S., Liu, Y., Sun, F., Cao, Q., & Shen, H. (2026). AuditAgent: LLM Agent for Risks Auditing in Recommender Systems. Proceedings of the AAAI Conference on Artificial Intelligence, 40(48), 41688-41690. https://doi.org/10.1609/aaai.v40i48.42382

Vancouver

Su D, Chen Z, Zhao S, Liu Y, Sun F, Cao Q, et al. AuditAgent: LLM Agent for Risks Auditing in Recommender Systems. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(48):41688-41690. doi:10.1609/aaai.v40i48.42382

BibTeX
@article{Su2026,
  author = {Su, Du and Chen, Zhenxing and Zhao, Shilong and Liu, Yuanhao and Sun, Fei and Cao, Qi and Shen, Huawei},
  title = {AuditAgent: LLM Agent for Risks Auditing in Recommender Systems},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {48},
  pages = {41688--41690},
  doi = {10.1609/aaai.v40i48.42382}
}
Kostenlos starten →