CLAug 16, 2025

LLM-as-a-Judge for Privacy Evaluation? Exploring the Alignment of Human and LLM Perceptions of Privacy in Textual Data

arXiv:2508.12158v111 citationsh-index: 7Proceedings of the 2025 Workshop on Human-Centered AI Privacy and Security
Originality Incremental advance
AI Analysis

This addresses the challenge of evaluating privacy in NLP, which is crucial for developing privacy-preserving technologies, though it appears incremental as it applies an existing LLM-as-a-Judge paradigm to a new domain.

The paper investigated whether large language models (LLMs) can effectively evaluate privacy sensitivity in textual data, finding that LLMs can accurately model a global human privacy perspective despite low inter-human agreement rates, based on a study with 10 datasets, 13 LLMs, and 677 human participants.

Despite advances in the field of privacy-preserving Natural Language Processing (NLP), a significant challenge remains the accurate evaluation of privacy. As a potential solution, using LLMs as a privacy evaluator presents a promising approach $\unicode{x2013}$ a strategy inspired by its success in other subfields of NLP. In particular, the so-called $\textit{LLM-as-a-Judge}$ paradigm has achieved impressive results on a variety of natural language evaluation tasks, demonstrating high agreement rates with human annotators. Recognizing that privacy is both subjective and difficult to define, we investigate whether LLM-as-a-Judge can also be leveraged to evaluate the privacy sensitivity of textual data. Furthermore, we measure how closely LLM evaluations align with human perceptions of privacy in text. Resulting from a study involving 10 datasets, 13 LLMs, and 677 human survey participants, we confirm that privacy is indeed a difficult concept to measure empirically, exhibited by generally low inter-human agreement rates. Nevertheless, we find that LLMs can accurately model a global human privacy perspective, and through an analysis of human and LLM reasoning patterns, we discuss the merits and limitations of LLM-as-a-Judge for privacy evaluation in textual data. Our findings pave the way for exploring the feasibility of LLMs as privacy evaluators, addressing a core challenge in solving pressing privacy issues with innovative technical solutions.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes