EUVIMEDEUVIMED entdecken
Crossref

Knowledge Boundary Discovery for Large Language Models

Ziquan Wang, Zhongqi Lu

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (40) · 33791-33799

DOI: 10.1609/aaai.v40i40.40670 ↗

Kurz gefasst

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

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We propose Knowledge Boundary Discovery (KBD), a reinforcement learning based framework to explore the knowledge boundaries of the Large Language Models (LLMs). We define the knowledge boundary by automatically generating two types of questions: (i) those the LLM can confidently answer within-knowledge boundary and (ii) those it cannot beyond-knowledge boundary. Iteratively exploring and exploiting the LLM's responses to find its knowledge boundaries is challenging because of the hallucination phenomenon. To find the knowledge boundaries of an LLM, the agent interacts with the LLM under the modeling of exploring a partially observable environment. The agent generates a progressive question as the action, adopts an entropy reduction as the reward, receives the LLM's response as the observation and updates its belief states. We demonstrate that the KBD detects knowledge boundaries of LLMs by automatically finding a set of non-trivial answerable and unanswerable questions. We validate the KBD by comparing its generated knowledge boundaries with manually crafted LLM benchmark datasets. Experiments show that our KBD-generated question set is comparable to the human-generated datasets. Our approach paves a new way to evaluate LLMs.

Zitieren

APA

Wang, Z., & Lu, Z. (2026). Knowledge Boundary Discovery for Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence, 40(40), 33791-33799. https://doi.org/10.1609/aaai.v40i40.40670

Vancouver

Wang Z, Lu Z. Knowledge Boundary Discovery for Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(40):33791-33799. doi:10.1609/aaai.v40i40.40670

BibTeX
@article{Wang2026,
  author = {Wang, Ziquan and Lu, Zhongqi},
  title = {Knowledge Boundary Discovery for Large Language Models},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {40},
  pages = {33791--33799},
  doi = {10.1609/aaai.v40i40.40670}
}
Kostenlos starten →