KBM: Delineating Knowledge Boundary for Adaptive Retrieval in Large Language Models

Zhen Zhang, Xinyu Wang, Yong Jiang, Zile Qiao, Zhuo Chen, Guangyu Li, Feiteng Mu, Mengting Hu, Pengjun Xie, Fei Huang


Abstract
Large Language Models (LLMs) often struggle with dynamically changing knowledge and handling unknown static information. Retrieval-Augmented Generation (RAG) is employed to tackle these challenges and has a significant impact on improving LLM performance. In fact, we find that not all questions need to trigger RAG. By retrieving parts of knowledge unknown to the LLM and allowing the LLM to answer the rest, we can effectively reduce both time and computational costs. In our work, we propose a Knowledge Boundary Model (KBM) to express the known/unknown of a given question, and to determine whether a RAG needs to be triggered. Experiments conducted on 11 English and Chinese datasets illustrate that the KBM effectively delineates the knowledge boundary, significantly decreasing the proportion of retrievals required for optimal end-to-end performance. Furthermore, we evaluate the effectiveness of KBM in three complex scenarios: dynamic knowledge, long-tail static knowledge, and multi-hop problems, as well as its functionality as an external LLM plug-in.
Anthology ID:
2025.findings-emnlp.1188
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2025
Month:
November
Year:
2025
Address:
Suzhou, China
Editors:
Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
21771–21782
Language:
URL:
https://aclanthology.org/2025.findings-emnlp.1188/
DOI:
Bibkey:
Cite (ACL):
Zhen Zhang, Xinyu Wang, Yong Jiang, Zile Qiao, Zhuo Chen, Guangyu Li, Feiteng Mu, Mengting Hu, Pengjun Xie, and Fei Huang. 2025. KBM: Delineating Knowledge Boundary for Adaptive Retrieval in Large Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 21771–21782, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
KBM: Delineating Knowledge Boundary for Adaptive Retrieval in Large Language Models (Zhang et al., Findings 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.findings-emnlp.1188.pdf
Checklist:
 2025.findings-emnlp.1188.checklist.pdf