Can LVLMs Uncover the Truth Behind Visual Illusions? An Analysis of Perceptual and Reasoning Capabilities

• URL: http://arxiv.org/abs/2607.27747v1
• Date: Thu, 30 Jul 2026 06:37:13 GMT
• Title: Can LVLMs Uncover the Truth Behind Visual Illusions? An Analysis of Perceptual and Reasoning Capabilities
• Abstract summary: We propose to evaluate Large Vision Language Models by exploiting visual illusions as a diagnostic tool.<n>Visual illusions are phenomena in which the human visual system misinterprets objective signals.<n>Based on IllusionReasoning, we show that the reasoning capabilities of a wide range of LVLMs are not as advanced as claimed.
• Score: 16.963382096640064
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Vision Language Models have integrated reasoning capabilities, elevating cognitive performance to new levels. However, existing evaluations either focus solely on perception or rely on specific domains such as maths or coding. Evaluation for reasoning capabilities that align with an open-world environment is still required, especially one that considers perception and reasoning jointly. To bridge this gap, we propose to evaluate LVLMs by exploiting visual illusions as a diagnostic tool. Visual illusions are phenomena in which the human visual system misinterprets objective signals, resulting in an understanding that deviates from reality. We constructed IllusionReasoning, a benchmark of illusion images collected from the real world, incorporating diverse annotated question-answer pairs. Based on IllusionReasoning, we show that the reasoning capabilities of a wide range of LVLMs are not as advanced as claimed. Our work provides new insights into LVLMs and offers future direction for optimisation.

Related papers

• Toward Cognitive Supersensing in Multimodal Large Language Model [67.15559571626747]
  We introduce Cognitive Supersensing, a training paradigm that endows MLLMs with human-like visual imagery capabilities.<n>In experiments, MLLMs trained with Cognitive Supersensing significantly outperform state-of-the-art baselines on CogSense-Bench.<n>We will open-source the CogSense-Bench and our model weights.
  arXiv  Detail & Related papers  (2026-02-02T02:19:50Z)
• ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs [98.27348724529257]
  We introduce ViCrit (Visual Caption Hallucination Critic), an RL proxy task that trains VLMs to localize a subtle, synthetic visual hallucination injected into paragraphs of human-written image captions.<n>Models trained with the ViCrit Task exhibit substantial gains across a variety of vision-language models benchmarks.
  arXiv  Detail & Related papers  (2025-06-11T19:16:54Z)
• Do Large Vision-Language Models Distinguish between the Actual and Apparent Features of Illusions? [12.157632635072435]
  Humans are susceptible to optical illusions, which serve as valuable tools for investigating sensory and cognitive processes.<n>Research has begun exploring whether machines, such as large vision language models (LVLMs), exhibit similar susceptibilities to visual illusions.
  arXiv  Detail & Related papers  (2025-06-06T05:47:50Z)
• Do you see what I see? An Ambiguous Optical Illusion Dataset exposing limitations of Explainable AI [4.58733012283457]
  We introduce a novel dataset of optical illusions featuring intermingled animal pairs designed to evoke perceptual ambiguity.<n>We identify generalizable visual concepts, particularly gaze direction and eye cues, as subtle yet impactful features that significantly influence model accuracy.<n>Our findings underscore the importance of concepts in visual learning and provide a foundation for studying bias and alignment between human and machine vision.
  arXiv  Detail & Related papers  (2025-05-27T12:22:59Z)
• Retrieval-Based Interleaved Visual Chain-of-Thought in Real-World Driving Scenarios [69.00444996464662]
  We propose RIV-CoT, a Retrieval-Based Interleaved Visual Chain-of-Thought method that enables vision-language models to reason using visual crops corresponding to relevant entities.<n>Our experiments demonstrate that RIV-CoT improves answer accuracy by 3.1% and reasoning accuracy by 4.6% over vanilla CoT prompting.
  arXiv  Detail & Related papers  (2025-01-08T18:31:16Z)
• IllusionBench+: A Large-scale and Comprehensive Benchmark for Visual Illusion Understanding in Vision-Language Models [56.34742191010987]
  Current Visual Language Models (VLMs) show impressive image understanding but struggle with visual illusions.<n>We introduce IllusionBench, a comprehensive visual illusion dataset that encompasses classic cognitive illusions and real-world scene illusions.<n>We design trap illusions that resemble classical patterns but differ in reality, highlighting issues in SOTA models.
  arXiv  Detail & Related papers  (2025-01-01T14:10:25Z)
• Visual Description Grounding Reduces Hallucinations and Boosts Reasoning in LVLMs [52.497823009176074]
  Large Vision-Language Models (LVLMs) often produce responses that misalign with factual information, a phenomenon known as hallucinations.<n>We introduce Visual Description Grounded Decoding (VDGD), a training-free method designed to enhance visual perception and improve reasoning capabilities in LVLMs.
  arXiv  Detail & Related papers  (2024-05-24T16:21:59Z)
• Mind's Eye of LLMs: Visualization-of-Thought Elicits Spatial Reasoning in Large Language Models [71.93366651585275]
  Large language models (LLMs) have exhibited impressive performance in language comprehension and various reasoning tasks. We propose Visualization-of-Thought (VoT) to elicit spatial reasoning of LLMs by visualizing their reasoning traces. VoT significantly enhances the spatial reasoning abilities of LLMs.
  arXiv  Detail & Related papers  (2024-04-04T17:45:08Z)
• IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models [21.589318022339317]
  We present IllusionVQA: a dataset of challenging optical illusions and hard-to-interpret scenes. Human evaluation reveals that humans achieve 91.03% and 100% accuracy in comprehension and localization.
  arXiv  Detail & Related papers  (2024-03-23T23:06:32Z)
• Analyzing and Mitigating Object Hallucination in Large Vision-Language Models [110.12460299261531]
  Large vision-language models (LVLMs) have shown remarkable abilities in understanding visual information with human languages. LVLMs still suffer from object hallucination, which is the problem of generating descriptions that include objects that do not actually exist in the images. We propose a powerful algorithm, LVLM Hallucination Revisor (LURE), to rectify object hallucination in LVLMs by reconstructing less hallucinatory descriptions.
  arXiv  Detail & Related papers  (2023-10-01T18:10:53Z)

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.