WikiVQABench: 一个基于维基百科和维基数据的知识引导视觉问答基准
WikiVQABench: A Knowledge-Grounded Visual Question Answering Benchmark from Wikipedia and Wikidata
AI总结：
本文提出WikiVQABench，一个结合维基百科图片、文章描述和维基数据结构化知识的知识引导视觉问答基准，通过大规模语言模型生成候选多选题，并由人工审核确保事实正确性和视觉-文本一致性，评估多种视觉-语言模型在知识密集型推理中的性能。
摘要：
视觉问答（VQA）基准大多强调基于感知的任务，这些任务可以通过单独的视觉内容解决。相比之下，许多现实场景需要外部知识来正确回答，而这些知识无法直接从图像中观察到。我们介绍了WikiVQABench，一个由系统结合维基百科图片、其相关文章描述和来自维基数据的结构化知识构建的人工整理的知识引导VQA基准。我们的流程使用大规模语言模型（LLMs）生成候选多选图像-问题-答案集。所有生成的实例随后由人工标注者审核，以确保事实正确性、视觉-文本一致性以及每个问题需要外部知识，除了视觉证据外，才能正确解决。WikiVQABench包含大量维基百科图片和经过整理的多选问题，旨在基准测试知识意识的视觉-语言模型（VLMs）。对十五种VLMs（256M-90B参数）的评估显示了广泛的性能范围（24.7%-75.6%准确率），表明该基准能够有效区分模型在知识密集型推理中的能力。数据集和基准测试代码已公开。
英文摘要：
Visual Question Answering (VQA) benchmarks have largely emphasized perception-based tasks that can be solved from visual content alone. In contrast, many real-world scenarios require external knowledge that is not directly observable in the image to answer correctly. We introduce WikiVQABench, a human-curated knowledge-grounded VQA benchmark constructed by systematically combining Wikipedia images, their associated article captions, and structured knowledge from Wikidata. Our pipeline uses large language models (LLMs) to generate candidate multiple-choice image-question-answer sets. All generated instances are subsequently reviewed and curated by human annotators to ensure factual correctness, visual-text consistency, and that each question requires external knowledge in addition to visual evidence for correct resolution. WikiVQABench comprises a substantial collection of Wikipedia images with curated multiple-choice questions designed to benchmark knowledge-aware vision-language models (VLMs). Evaluation of fifteen VLMs (256M-90B parameters) reveals a wide performance range (24.7%-75.6% accuracy), demonstrating that the benchmark effectively discriminates model capabilities on knowledge-intensive reasoning. The dataset and benchmarking code are publicly available.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
