VLAgeBench: Benchmarking Large Vision-Language Models for Zero-Shot Human Age Estimation

2026-03-26

AI Summary

This study investigates the feasibility of large vision-language models (LVLMs) for face age estimation under zero-shot settings without labeled training data. We systematically evaluate state-of-the-art models—including GPT-4o, Claude 3.5 Sonnet, and LLaMA 3.2 Vision—on the UTKFace and FG-NET datasets using eight metrics such as mean absolute error (MAE), coefficient of determination (R²), concordance correlation coefficient (CCC), and accuracy within ±5 years. Our analysis further examines the impact of image quality and demographic subgroups. We establish the first reproducible benchmark for zero-shot face age estimation, revealing both the promising capabilities and fairness challenges of general-purpose multimodal models in biometric applications. Results demonstrate that LVLMs can achieve competitive performance in zero-shot scenarios, yet remain constrained by prompt sensitivity, computational cost, and disparities across demographic groups.

Abstract

Human age estimation from facial images represents a challenging computer vision task with significant applications in biometrics, healthcare, and human-computer interaction. While traditional deep learning approaches require extensive labeled datasets and domain-specific training, recent advances in large vision-language models (LVLMs) offer the potential for zero-shot age estimation. This study presents a comprehensive zero-shot evaluation of state-of-the-art Large Vision-Language Models (LVLMs) for facial age estimation, a task traditionally dominated by domain-specific convolutional networks and supervised learning. We assess the performance of GPT-4o, Claude 3.5 Sonnet, and LLaMA 3.2 Vision on two benchmark datasets, UTKFace and FG-NET, without any fine-tuning or task-specific adaptation. Using eight evaluation metrics, including MAE, MSE, RMSE, MAPE, MBE, $R^2$, CCC, and $\pm$5-year accuracy, we demonstrate that general-purpose LVLMs can deliver competitive performance in zero-shot settings. Our findings highlight the emergent capabilities of LVLMs for accurate biometric age estimation and position these models as promising tools for real-world applications. Additionally, we highlight performance disparities linked to image quality and demographic subgroups, underscoring the need for fairness-aware multimodal inference. This work introduces a reproducible benchmark and positions LVLMs as promising tools for real-world applications in forensic science, healthcare monitoring, and human-computer interaction. The benchmark focuses on strict zero-shot inference without fine-tuning and highlights remaining challenges related to prompt sensitivity, interpretability, computational cost, and demographic fairness.

Problem

• zero-shot age estimation
• large vision-language models
• facial age estimation
• demographic fairness
• biometric inference

Innovation

• zero-shot age estimation
• large vision-language models
• biometric fairness
• VLAgeBench
• multimodal inference