FrequencyQA: Evaluating Factuality Robustness of LLMs Through Knowledge Frequency Control and Uncertainty Estimation

ACL ARR 2025 February Submission4528 Authors

Abstract:

The rapid development of LLMs has sparked extensive research into their factual knowledge. Current works find that LLMs fall short on questions around low-frequency entities. However, such proofs are unreliable since the questions can differ not only in entity frequency but also in difficulty themselves. So we introduce **FrequencyQA** benchmark, containing **283K** abstract questions, each instantiated by a pair of high-frequency and low-frequency entities. It ensures a controllable comparison to study the role of knowledge frequency in the performance of LLMs. Because the difference between such a pair is only the entity with different frequencies.
In addition, we use both correctness and uncertainty to develop a two-round method to evaluate LLMs' knowledge robustness. It aims to avoid possible semantic shortcuts which is a serious problem of current QA study. Experiments reveal that LLMs, including GPT-4o, exhibit particularly low robustness regarding low-frequency knowledge. Besides, we find that uncertainty can be used to effectively identify high-quality and shortcut-free questions while maintaining the data size. Based on this, we propose an automatic method to select such questions to form a subset called **FrequencyQA-Hard**, containing only hard low-frequency questions.

Paper Type: Long

Research Area: Resources and Evaluation

Research Area Keywords: Benchmarking, NLP datasets, Evaluation Methodologies

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Data resources, Data analysis

Languages Studied: English

Submission Number: 4528