---
format: "aidr-story-markdown/v1"
id: "6a73c142798d871224e62f341be249aab40b7eeb3451be3bc79a7001198a7909"
canonical_url: "https://aidr.today/6a73c142?lang=en"
title: "MInfoVisQA: A Multilingual Benchmark for Information-Rich Visual Question Answering"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-18T05:00:00.000Z"
category: "Data"
topics: ["llm","data-engineering"]
source_urls: ["https://arxiv.org/abs/2611.21644"]
summary: "Multimodal large language models (MLLMs) excel in vision-language tasks; yet, whether they maintain consistent understanding of information-rich visuals (InfoVis) when the surface language changes remains unclear. To quantify this capability, we introduce MInfoVisQA, a parallel benchmark with 8,960 QA configurations derived from 128 seed instances across 24 languages. We construct seed-matched, language-parallel visuals and QA pairs by disentangling numerical-structural data from text and programmatically synchronizing translated labels. We evaluate 25 MLLMs under two settings: Language-Aligned QA (LQA, matching visual-text languages) and Cross-Language QA (XQA, fixed QA language with varying visual languages). The results show that aggregate accuracy masks substantial robustness gaps: two models with comparable mean LQA accuracy (80.4\\%) exhibit performance spreads of 7.8 and 20.3 percentage points across languages. Moreover, all 25 models score lower with non-English visuals under fixed English QA, underscoring the need for cross-language sensitivity metrics."
---

# MInfoVisQA: A Multilingual Benchmark for Information\-Rich Visual Question Answering

> [Open the canonical story](<https://aidr.today/6a73c142?lang=en>)

**Published:** 2026-11-18T05:00:00.000Z
**Category:** Data
**Topics:** llm, data\-engineering

## Summary

Multimodal large language models \(MLLMs\) excel in vision\-language tasks; yet, whether they maintain consistent understanding of information\-rich visuals \(InfoVis\) when the surface language changes remains unclear\. To quantify this capability, we introduce MInfoVisQA, a parallel benchmark with 8,960 QA configurations derived from 128 seed instances across 24 languages\. We construct seed\-matched, language\-parallel visuals and QA pairs by disentangling numerical\-structural data from text and programmatically synchronizing translated labels\. We evaluate 25 MLLMs under two settings: Language\-Aligned QA \(LQA, matching visual\-text languages\) and Cross\-Language QA \(XQA, fixed QA language with varying visual languages\)\. The results show that aggregate accuracy masks substantial robustness gaps: two models with comparable mean LQA accuracy \(80\.4\\%\) exhibit performance spreads of 7\.8 and 20\.3 percentage points across languages\. Moreover, all 25 models score lower with non\-English visuals under fixed English QA, underscoring the need for cross\-language sensitivity metrics\.

## Sources

- [Story source](<https://arxiv.org/abs/2611.21644>)

