SU-FMI-AI/ImageCLEF-MR2026-MCQ-Textual

---
license: cc-by-nc-4.0
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
dataset_info:
features:
- name: question_id
dtype: string
- name: type
dtype: string
- name: visual_elements
struct:
- name: diagram
dtype: int64
- name: graph
dtype: int64
- name: other_figure
dtype: int64
- name: tables
dtype: int64
- name: language
dtype: string
- name: subject
dtype: string
- name: image
dtype: image
splits:
- name: test
num_bytes: 133095598
num_examples: 1653
download_size: 138216000
dataset_size: 133095598
---

# ImageCLEF 2026 Multimodal Reasoning Multiple-Choice QA Dataset - Textual questions

## Introduction
The test split is newly collected and annotated real-world data.
Participants are expected to send their model predictions to the leaderboard on [AI4Media-Bench MCQ Textual](https://ai4media-bench.aimultimedialab.ro/competitions/16/). The `test` split for the final leaderboard is available.

## Test Distribution

### Language Distribution
The test split covers 6 languages, distributed as follows:

| Language | Count |
| :--- | :--- |
| Bulgarian | 451 |
| Chinese | 267 |
| Croatian | 335 |
| English | 240 |
| Italian | 173 |
| Serbian | 187 |
| **Total** | **1653** |