---
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: answer_key
    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: 186026943
    num_examples: 1117
  download_size: 190318349
  dataset_size: 186026943
---

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

## Introduction
The data is newly collected and annotated real-world data. 
Participants are expected to send their model predictions to the leaderboard |on [AI4Media-Bench MCQ Visual](https://ai4media-bench.aimultimedialab.ro/competitions/12/). The `test` split for the final leaderboard is available.

## Test Distribution

### Language Distribution
The test split is distributed across several languages as follows:

| Language | Count |
| :--- | :---: |
| Bulgarian | 110 |
| Chinese | 342 |
| Croatian | 57 |
| English | 500 |
| Italian | 54 |
| Serbian | 54 |
| **Total** | **1117** |

### Graphical Distribution
The image-based test split contains various graphical elements:

| Graphical Feature | Count | Percentage |
| :--- | :---: | :---: |
| Diagrams | 644 | 57.65% |
| Graphs | 177 | 15.85% |
| Tables | 298 | 26.68% |
| Other Figures | 129 | 11.55% |