Auditing Disability Representation in Vision-Language Models

• URL: http://arxiv.org/abs/2601.17348v1
• Date: Sat, 24 Jan 2026 07:25:43 GMT
• Title: Auditing Disability Representation in Vision-Language Models
• Authors: Srikant Panda, Sourabh Singh Yadav, Palkesh Malviya,
• Abstract summary: We study disability aware descriptions for person centric images.<n>We introduce a benchmark based on paired Neutral Prompts (NP) and Disability-Contextualised Prompts (DP)<n>We evaluate 15 state-of-the-art open- and closed-source vision-language models under a zero-shot setting across 9 disability categories.
• Score: 0.6987503477818553
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-language models (VLMs) are increasingly deployed in socially sensitive applications, yet their behavior with respect to disability remains underexplored. We study disability aware descriptions for person centric images, where models often transition from evidence grounded factual description to interpretation shift including introduction of unsupported inferences beyond observable visual evidence. To systematically analyze this phenomenon, we introduce a benchmark based on paired Neutral Prompts (NP) and Disability-Contextualised Prompts (DP) and evaluate 15 state-of-the-art open- and closed-source VLMs under a zero-shot setting across 9 disability categories. Our evaluation framework treats interpretive fidelity as core objective and combines standard text-based metrics capturing affective degradation through shifts in sentiment, social regard and response length with an LLM-as-judge protocol, validated by annotators with lived experience of disability. We find that introducing disability context consistently degrades interpretive fidelity, inducing interpretation shifts characterised by speculative inference, narrative elaboration, affective degradation and deficit oriented framing. These effects are further amplified along race and gender dimension. Finally, we demonstrate targeted prompting and preference fine-tuning effectively improves interpretive fidelity and reduces substantially interpretation shifts.

Related papers

• Investigating Disability Representations in Text-to-Image Models [7.244686394468418]
  This study investigates how people with disabilities are represented in AI-generated images.<n>We analyze disability representations by comparing image similarities between generic disability prompts and prompts referring to specific disability categories.
  arXiv  Detail & Related papers  (2026-02-04T15:54:25Z)
• ReViP: Reducing False Completion in Vision-Language-Action Models with Vision-Proprioception Rebalance [50.05984919728878]
  We present ReViP, a novel VLA framework with Vision-Proprioception Rebalance to enhance visual grounding and robustness under perturbations.<n>Specifically, we use an external VLM as a task-stage observer to extract real-time task-centric visual cues from visual observations.<n>To evaluate false completion, we propose the first False-Completion Benchmark Suite built on LIBERO with controlled settings such as Object-Drop.
  arXiv  Detail & Related papers  (2026-01-23T11:31:07Z)
• Stable Language Guidance for Vision-Language-Action Models [62.80963701282789]
  Residual Semantic Steering is a probabilistic framework that disentangles physical affordance from semantic execution.<n> RSS achieves state-of-the-art robustness, maintaining performance even under adversarial linguistic perturbations.
  arXiv  Detail & Related papers  (2026-01-07T16:16:10Z)
• AccessEval: Benchmarking Disability Bias in Large Language Models [3.160274015679566]
  Large Language Models (LLMs) are increasingly deployed across diverse domains but often exhibit disparities in how they handle real-life queries.<n>We introduce textbfAccessEval (Accessibility Evaluation), a benchmark evaluating 21 closed- and open-source LLMs across 6 real-world domains and 9 disability types.<n>Our analysis reveals that responses to disability-aware queries tend to have a more negative tone, increased stereotyping, and higher factual error compared to neutral queries.
  arXiv  Detail & Related papers  (2025-09-22T17:49:03Z)
• Who's Asking? Investigating Bias Through the Lens of Disability Framed Queries in LLMs [2.722784054643991]
  Large Language Models (LLMs) routinely infer users demographic traits from phrasing alone.<n>Disability cues in shaping these inferences remains largely uncharted.<n>We present the first systematic audit of disability-conditioned demographic bias across eight state-of-the-art instruction-tuned LLMs.
  arXiv  Detail & Related papers  (2025-08-18T21:03:09Z)
• SHALE: A Scalable Benchmark for Fine-grained Hallucination Evaluation in LVLMs [52.03164192840023]
  Large Vision-Language Models (LVLMs) still suffer from hallucinations, i.e., generating content inconsistent with input or established world knowledge.<n>We propose an automated data construction pipeline that produces scalable, controllable, and diverse evaluation data.<n>We construct SHALE, a benchmark designed to assess both faithfulness and factuality hallucinations.
  arXiv  Detail & Related papers  (2025-08-13T07:58:01Z)
• Reasoning Beyond Labels: Measuring LLM Sentiment in Low-Resource, Culturally Nuanced Contexts [10.492471013369782]
  We present a framework that treats sentiment as a context-dependent, culturally embedded construct.<n>We evaluate how large language models (LLMs) reason about sentiment in WhatsApp messages from Nairobi youth health groups.
  arXiv  Detail & Related papers  (2025-08-06T08:27:55Z)
• Estimating Commonsense Plausibility through Semantic Shifts [66.06254418551737]
  We propose ComPaSS, a novel discriminative framework that quantifies commonsense plausibility by measuring semantic shifts.<n> Evaluations on two types of fine-grained commonsense plausibility estimation tasks show that ComPaSS consistently outperforms baselines.
  arXiv  Detail & Related papers  (2025-02-19T06:31:06Z)
• Sycophancy in Vision-Language Models: A Systematic Analysis and an Inference-Time Mitigation Framework [18.54098084470481]
  We analyze sycophancy across vision-language benchmarks and propose an inference-time mitigation framework.<n>Our framework effectively mitigates sycophancy across all evaluated models, while maintaining performance on neutral prompts.
  arXiv  Detail & Related papers  (2024-08-21T01:03:21Z)
• 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)
• Simple Linguistic Inferences of Large Language Models (LLMs): Blind Spots and Blinds [59.71218039095155]
  We evaluate language understanding capacities on simple inference tasks that most humans find trivial. We target (i) grammatically-specified entailments, (ii) premises with evidential adverbs of uncertainty, and (iii) monotonicity entailments. The models exhibit moderate to low performance on these evaluation sets.
  arXiv  Detail & Related papers  (2023-05-24T06:41:09Z)

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.