Do Large Language Models Know How Much They Know?

2025-02-26

AI Summary

This study investigates whether large language models (LLMs) possess metacognitive awareness of their own knowledge boundaries—specifically, the ability to accurately assess “what they know versus what they do not know.” Method: We introduce the first systematic knowledge enumeration benchmark, requiring models to exhaustively list factual statements about a given topic; accuracy of self-assessed knowledge volume is quantified via human verification and automated consistency scoring. Our approach employs prompt engineering and controllable generation to instantiate knowledge enumeration. Results: Knowledge-awareness emerges with increasing parameter count and is robust across diverse LLM architectures. All evaluated state-of-the-art LLMs significantly distinguish between under-, over-, and appropriately informative outputs. This provides the first empirical evidence that LLMs exhibit rudimentary yet measurable capability to identify their knowledge boundaries—establishing a novel paradigm for trustworthy model evaluation and controllable text generation.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Large Language Models (LLMs) have emerged as highly capable systems and are increasingly being integrated into various uses. However, the rapid pace of their deployment has outpaced a comprehensive understanding of their internal mechanisms and a delineation of their capabilities and limitations. A desired attribute of an intelligent system is its ability to recognize the scope of its own knowledge. To investigate whether LLMs embody this characteristic, we develop a benchmark designed to challenge these models to enumerate all information they possess on specific topics. This benchmark evaluates whether the models recall excessive, insufficient, or the precise amount of information, thereby indicating their awareness of their own knowledge. Our findings reveal that all tested LLMs, given sufficient scale, demonstrate an understanding of how much they know about specific topics. While different architectures exhibit varying rates of this capability's emergence, the results suggest that awareness of knowledge may be a generalizable attribute of LLMs. Further research is needed to confirm this potential and fully elucidate the underlying mechanisms.

Problem

• LLMs knowledge self-awareness
• benchmark for LLMs knowledge evaluation
• LLMs capability and limitation delineation

Innovation

• Developed benchmark for LLMs
• Evaluated LLMs' knowledge awareness
• Tested LLMs' information recall accuracy