EUVIMEDEUVIMED entdecken
Crossref

Fine-Grained Interpretation of Political Opinions in Large Language Models

Jingyu Hu, Mengyue Yang, Mengnan Du, Weiru Liu

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (45) · 38570-38579

DOI: 10.1609/aaai.v40i45.41199 ↗

Kurz gefasst

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

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Studies of LLMs’ political opinions mainly evaluate their open-ended responses. Recent work indicates misalignment between LLMs responses and their internal intentions. This motivates us to probe LLMs' internal mechanisms and uncover their internal political states. Additionally, analysis of LLMs' political opinions often relies on single-axis concepts, which can lead to concept confounds. Our work extends this to multi-dimensions and applies interpretable techniques for more transparent LLM political concept learning. Specifically, we designed a four-dimensional political learning framework and constructed a corresponding dataset for fine-grained political concept vector learning. These vectors can detect and intervene in LLM internals. Experiments are conducted on eight open-source LLMs with three representation engineering techniques. Results show these vectors can disentangle political concept confounds. Detection tasks validate the semantic meaning of the vectors and show good generalization and robustness in OOD settings. Intervention experiments show that these vectors can implicitly intervene in LLMs, generating responses with targeted political leanings. These insights reveal the need for more transparent auditing for future AI governance.

Zitieren

APA

Hu, J., Yang, M., Du, M., & Liu, W. (2026). Fine-Grained Interpretation of Political Opinions in Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence, 40(45), 38570-38579. https://doi.org/10.1609/aaai.v40i45.41199

Vancouver

Hu J, Yang M, Du M, Liu W. Fine-Grained Interpretation of Political Opinions in Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(45):38570-38579. doi:10.1609/aaai.v40i45.41199

BibTeX
@article{Hu2026,
  author = {Hu, Jingyu and Yang, Mengyue and Du, Mengnan and Liu, Weiru},
  title = {Fine-Grained Interpretation of Political Opinions in Large Language Models},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {45},
  pages = {38570--38579},
  doi = {10.1609/aaai.v40i45.41199}
}
Kostenlos starten →