arXiv 2603.154092026-03-16cs.CL

SEA-Vision：面向东南亚的多语言综合文档和场景文本理解基准

SEA-Vision: A Multilingual Benchmark for Comprehensive Document and Scene Text Understanding in Southeast Asia

Xiamen University, China（厦门大学，中国）；Shopee, China（Shopee，中国）；Tongji University, China（同济大学，中国）

Pengfei Yue, Xingran Zhao, Juntao Chen, Peng Hou, Wang Longchao, Jianghang Lin, Shengchuan Zhang, Anxiang Zeng, Liujuan Cao

备注：: Accepted By CVPR2026

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

SEA-Vision提出一个多语言基准，用于评估文档解析和文本中心视觉问答，涵盖11种东南亚语言，包含15234页文档和7496对问答对，揭示多语言文档理解的差距。

摘要：

多语言文档和场景文本理解在搜索、金融和公共服务中具有重要作用。然而，现有基准多关注高资源语言，无法评估现实多语言环境中的模型。在东南亚，语言多样性、复杂书写系统和多样的文档类型使这一挑战更加严峻。我们引入SEA-Vision基准，联合评估文档解析和文本中心视觉问答（TEC-VQA）跨11种东南亚语言。SEA-Vision包含15,234页文档解析页面，标注有分层的页面、块和行级别标签。它还提供7,496对TEC-VQA问题-答案对，测试文本识别、数值计算、比较分析、逻辑推理和空间理解。为使多语言、多任务标注可行，我们设计了一种混合流水线，结合自动过滤和评分、MLLM辅助标注和轻量级母语者验证，大幅减少人工标注的同时保持高质量。我们评估了多个领先的多模态模型，并观察到在低资源东南亚语言上性能显著下降，突显了多语言文档和场景文本理解存在巨大差距。我们相信SEA-Vision将推动全球文档和场景文本理解的进步。

英文摘要：

Multilingual document and scene text understanding plays an important role in applications such as search, finance, and public services. However, most existing benchmarks focus on high-resource languages and fail to evaluate models in realistic multilingual environments. In Southeast Asia, the diversity of languages, complex writing systems, and highly varied document types make this challenge even greater. We introduce SEA-Vision, a benchmark that jointly evaluates Document Parsing and Text-Centric Visual Question Answering (TEC-VQA) across 11 Southeast Asian languages. SEA-Vision contains 15,234 document parsing pages from nine representative document types, annotated with hierarchical page-, block-, and line-level labels. It also provides 7,496 TEC-VQA question-answer pairs that probe text recognition, numerical calculation, comparative analysis, logical reasoning, and spatial understanding. To make such multilingual, multi-task annotation feasible, we design a hybrid pipeline for Document Parsing and TEC-VQA. It combines automated filtering and scoring with MLLM-assisted labeling and lightweight native-speaker verification, greatly reducing manual labeling while maintaining high quality. We evaluate several leading multimodal models and observe pronounced performance degradation on low-resource Southeast Asian languages, highlighting substantial remaining gaps in multilingual document and scene text understanding. We believe SEA-Vision will help drive global progress in document and scene text understanding.

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。