DemoBias: An Empirical Study to Trace Demographic Biases in Vision Foundation Models

• URL: http://arxiv.org/abs/2508.19298v1
• Date: Mon, 25 Aug 2025 18:02:49 GMT
• Title: DemoBias: An Empirical Study to Trace Demographic Biases in Vision Foundation Models
• Authors: Abu Sufian, Anirudha Ghosh, Debaditya Barman, Marco Leo, Cosimo Distante,
• Abstract summary: Large Vision Language Models (LVLMs) have demonstrated remarkable capabilities across various downstream tasks, including biometric face recognition (FR) with description.<n>We conduct an empirical evaluation to investigate the extent of demographic biases in LVLMs for biometric FR with textual token generation tasks.
• Score: 5.024921806058944
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Large Vision Language Models (LVLMs) have demonstrated remarkable capabilities across various downstream tasks, including biometric face recognition (FR) with description. However, demographic biases remain a critical concern in FR, as these foundation models often fail to perform equitably across diverse demographic groups, considering ethnicity/race, gender, and age. Therefore, through our work DemoBias, we conduct an empirical evaluation to investigate the extent of demographic biases in LVLMs for biometric FR with textual token generation tasks. We fine-tuned and evaluated three widely used pre-trained LVLMs: LLaVA, BLIP-2, and PaliGemma on our own generated demographic-balanced dataset. We utilize several evaluation metrics, like group-specific BERTScores and the Fairness Discrepancy Rate, to quantify and trace the performance disparities. The experimental results deliver compelling insights into the fairness and reliability of LVLMs across diverse demographic groups. Our empirical study uncovered demographic biases in LVLMs, with PaliGemma and LLaVA exhibiting higher disparities for Hispanic/Latino, Caucasian, and South Asian groups, whereas BLIP-2 demonstrated comparably consistent. Repository: https://github.com/Sufianlab/DemoBias.

Related papers

• Mitigating Group-Level Fairness Disparities in Federated Visual Language Models [115.16940773660104]
  This paper introduces FVL-FP, a novel framework that combines FL with fair prompt tuning techniques.<n>We focus on mitigating demographic biases while preserving model performance.<n>Our approach reduces demographic disparity by an average of 45% compared to standard FL approaches.
  arXiv  Detail & Related papers  (2025-05-03T16:09:52Z)
• BiasICL: In-Context Learning and Demographic Biases of Vision Language Models [0.7499722271664147]
  Vision language models (VLMs) show promise in medical diagnosis, but their performance across demographic subgroups when using in-context learning (ICL) remains poorly understood.<n>We examine how the demographic composition of demonstration examples affects VLM performance in two medical imaging tasks: skin lesion malignancy prediction and pneumothorax detection from chest radiographs.
  arXiv  Detail & Related papers  (2025-03-04T06:45:54Z)
• Human Cognitive Benchmarks Reveal Foundational Visual Gaps in MLLMs [65.93003087656754]
  VisFactor is a benchmark that digitizes 20 vision-centric subtests from a well-established cognitive psychology assessment.<n>We evaluate 20 frontier Multimodal Large Language Models (MLLMs) from GPT, Gemini, Claude, LLaMA, Qwen, and SEED families.<n>The best-performing model achieves a score of only 25.19 out of 100, with consistent failures on tasks such as mental rotation, spatial relation inference, and figure-ground discrimination.
  arXiv  Detail & Related papers  (2025-02-23T04:21:32Z)
• CEB: Compositional Evaluation Benchmark for Fairness in Large Language Models [58.57987316300529]
  Large Language Models (LLMs) are increasingly deployed to handle various natural language processing (NLP) tasks.<n>To evaluate the biases exhibited by LLMs, researchers have recently proposed a variety of datasets.<n>We propose CEB, a Compositional Evaluation Benchmark that covers different types of bias across different social groups and tasks.
  arXiv  Detail & Related papers  (2024-07-02T16:31:37Z)
• GenderBias-\emph{VL}: Benchmarking Gender Bias in Vision Language Models via Counterfactual Probing [72.0343083866144]
  This paper introduces the GenderBias-emphVL benchmark to evaluate occupation-related gender bias in Large Vision-Language Models. Using our benchmark, we extensively evaluate 15 commonly used open-source LVLMs and state-of-the-art commercial APIs. Our findings reveal widespread gender biases in existing LVLMs.
  arXiv  Detail & Related papers  (2024-06-30T05:55:15Z)
• Cross-Care: Assessing the Healthcare Implications of Pre-training Data on Language Model Bias [3.455189439319919]
  We introduce Cross-Care, the first benchmark framework dedicated to assessing biases and real world knowledge in large language models (LLMs) We evaluate how demographic biases embedded in pre-training corpora like $ThePile$ influence the outputs of LLMs. Our results highlight substantial misalignment between LLM representation of disease prevalence and real disease prevalence rates across demographic subgroups.
  arXiv  Detail & Related papers  (2024-05-09T02:33:14Z)
• Seeds of Stereotypes: A Large-Scale Textual Analysis of Race and Gender Associations with Diseases in Online Sources [1.8259644946867188]
  The study analyzed the context in which various diseases are discussed alongside markers of race and gender. We found that demographic terms are disproportionately associated with specific disease concepts in online texts. We find widespread disparities in the associations of specific racial and gender terms with the 18 diseases analyzed.
  arXiv  Detail & Related papers  (2024-05-08T13:38:56Z)
• VALOR-EVAL: Holistic Coverage and Faithfulness Evaluation of Large Vision-Language Models [57.43276586087863]
  Large Vision-Language Models (LVLMs) suffer from hallucination issues, wherein the models generate plausible-sounding but factually incorrect outputs. Existing benchmarks are often limited in scope, focusing mainly on object hallucinations. We introduce a multi-dimensional benchmark covering objects, attributes, and relations, with challenging images selected based on associative biases.
  arXiv  Detail & Related papers  (2024-04-22T04:49:22Z)
• Sociodemographic Prompting is Not Yet an Effective Approach for Simulating Subjective Judgments with LLMs [13.744746481528711]
  Large Language Models (LLMs) are widely used to simulate human responses across diverse contexts.<n>We evaluate nine popular LLMs on their ability to understand demographic differences in two subjective judgment tasks: politeness and offensiveness.<n>We find that in zero-shot settings, most models' predictions for both tasks align more closely with labels from White participants than those from Asian or Black participants.
  arXiv  Detail & Related papers  (2023-11-16T10:02:24Z)
• Balancing Biases and Preserving Privacy on Balanced Faces in the Wild [50.915684171879036]
  There are demographic biases present in current facial recognition (FR) models. We introduce our Balanced Faces in the Wild dataset to measure these biases across different ethnic and gender subgroups. We find that relying on a single score threshold to differentiate between genuine and imposters sample pairs leads to suboptimal results. We propose a novel domain adaptation learning scheme that uses facial features extracted from state-of-the-art neural networks.
  arXiv  Detail & Related papers  (2021-03-16T15:05:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.