HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models

• URL: http://arxiv.org/abs/2310.14566v5
• Date: Mon, 25 Mar 2024 06:05:24 GMT
• Title: HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models
• Authors: Tianrui Guan, Fuxiao Liu, Xiyang Wu, Ruiqi Xian, Zongxia Li, Xiaoyu Liu, Xijun Wang, Lichang Chen, Furong Huang, Yaser Yacoob, Dinesh Manocha, Tianyi Zhou,
• Abstract summary: We introduce HallusionBench, a benchmark for the evaluation of image-context reasoning. The benchmark comprises 346 images paired with 1129 questions, all meticulously crafted by human experts. In our evaluation on HallusionBench, we benchmarked 15 different models, highlighting a 31.42% question-pair accuracy achieved by the state-of-the-art GPT-4V.
• Score: 69.52245481329899
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We introduce HallusionBench, a comprehensive benchmark designed for the evaluation of image-context reasoning. This benchmark presents significant challenges to advanced large visual-language models (LVLMs), such as GPT-4V(Vision), Gemini Pro Vision, Claude 3, and LLaVA-1.5, by emphasizing nuanced understanding and interpretation of visual data. The benchmark comprises 346 images paired with 1129 questions, all meticulously crafted by human experts. We introduce a novel structure for these visual questions designed to establish control groups. This structure enables us to conduct a quantitative analysis of the models' response tendencies, logical consistency, and various failure modes. In our evaluation on HallusionBench, we benchmarked 15 different models, highlighting a 31.42% question-pair accuracy achieved by the state-of-the-art GPT-4V. Notably, all other evaluated models achieve accuracy below 16%. Moreover, our analysis not only highlights the observed failure modes, including language hallucination and visual illusion, but also deepens an understanding of these pitfalls. Our comprehensive case studies within HallusionBench shed light on the challenges of hallucination and illusion in LVLMs. Based on these insights, we suggest potential pathways for their future improvement. The benchmark and codebase can be accessed at https://github.com/tianyi-lab/HallusionBench.

Related papers

• DRIVINGVQA: Analyzing Visual Chain-of-Thought Reasoning of Vision Language Models in Real-World Scenarios with Driving Theory Tests [69.00444996464662]
  We present DrivingVQA, a new benchmark derived from driving theory tests to evaluate visual chain-of-thought reasoning in complex real-world scenarios. Our experiments reveal that open-source and proprietary LVLMs struggle with visual chain-of-thought reasoning under zero-shot settings. We investigate training strategies that leverage relevant entities to improve visual reasoning.
  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. We introduce IllusionBench, a comprehensive visual illusion dataset that encompasses classic cognitive illusions and real-world scene illusions. 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)
• VisOnlyQA: Large Vision Language Models Still Struggle with Visual Perception of Geometric Information [9.420776624656144]
  We introduce VisOnlyQA, a new dataset to evaluate the visual perception capabilities of Large Vision Language Models (LVLMs) Our dataset enables us to analyze the visual perception of LVLMs for fine-grained visual information, independent of other capabilities such as reasoning.
  arXiv  Detail & Related papers  (2024-12-01T19:46:22Z)
• Explore the Hallucination on Low-level Perception for MLLMs [83.12180878559295]
  We aim to define and evaluate the self-awareness of MLLMs in low-level visual perception and understanding tasks. We present QL-Bench, a benchmark settings to simulate human responses to low-level vision. We demonstrate that while some models exhibit robust low-level visual capabilities, their self-awareness remains relatively underdeveloped.
  arXiv  Detail & Related papers  (2024-09-15T14:38:29Z)
• AutoHallusion: Automatic Generation of Hallucination Benchmarks for Vision-Language Models [91.78328878860003]
  Large vision-language models (LVLMs) are prone to hallucinations. benchmarks often rely on hand-crafted corner cases whose failure patterns may not generalize well. We develop AutoHallusion, the first automated benchmark generation approach.
  arXiv  Detail & Related papers  (2024-06-16T11:44:43Z)
• Mitigating Hallucination in Visual Language Models with Visual Supervision [33.05550629039951]
  Large vision-language models (LVLMs) suffer from hallucination a lot. Key problem lies in its weak ability to comprehend detailed content in a multi-modal context. In this paper, we bring more detailed vision annotations and more discriminative vision models to facilitate the training of LVLMs.
  arXiv  Detail & Related papers  (2023-11-27T09:30:02Z)
• Holistic Analysis of Hallucination in GPT-4V(ision): Bias and Interference Challenges [54.42256219010956]
  This benchmark is designed to evaluate and shed light on the two common types of hallucinations in visual language models: bias and interference. bias refers to the model's tendency to hallucinate certain types of responses, possibly due to imbalance in its training data. interference pertains to scenarios where the judgment of GPT-4V(ision) can be disrupted due to how the text prompt is phrased or how the input image is presented.
  arXiv  Detail & Related papers  (2023-11-06T17:26:59Z)

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.