patriciavnencakova/SlovakBebrasChallenge

---
language:
- sk
pretty_name: iBobor Slovak Questions Dataset
task_categories:
- question-answering
- multiple-choice
tags:
- education
- slovak
- bebras
- ibobor
- multiple-choice
size_categories:
- n<1K
---

# iBobor Slovak Questions Dataset

## Dataset Description

This dataset contains questions from the Slovak **iBobor** competition, publicly available on the website http://demo.ibobor.sk/sutaz_demo/, which is part of the international **Bebras Challenge** focused on informatics and computational thinking.

The dataset was created for the purposes of a master's thesis at **Comenius University in Bratislava, Faculty of Mathematics, Physics and Informatics**. The thesis is concerned with testing and evaluating large multimodal language models (LLMs) on Slovak logical, reasoning, and problem-solving tasks.

Each record represents one competition task and may include:
- task metadata,
- question text,
- answer choices,
- correct answer,
- and image URLs if the task contains visual content.

The dataset is intended to support research on the capabilities of modern LLMs in solving Slovak multiple-choice tasks, including both text-based and visually grounded questions.

## Languages

The dataset is in **Slovak**.

## Dataset Structure

At the moment, the dataset is provided as a JSON file with the task records.

Example fields:
| Key | Type | Description |
|---|---|---|
| `id` | integer | Unique identifier assigned to each task. |
| `serial_number` | integer | The task number within a particular competition test. |
| `title` | string | Title of the task. |
| `question` | string | Task statement, where embedded images are represented by placeholders such as `<image_1>`; each placeholder refers to the image at the corresponding position in the `images` array. |
| `choices` | array<object> | Array of answer choices, where each object has the form `{"value": "...", "label": "..."}`. The `value` field stores the option identifier, while `label` contains the answer content. In image-based tasks, `label` may include placeholders such as `<image_1>`, referring to the corresponding position in the `images` array. |
| `images` | array<string> | Array of image references associated with the task. |
| `year` | string | Competition year of the task. |
| `category` | string | Competition category of the task. |
| `type` | string | Detected task type (e.g. text-based). |
| `correct_index` | integer | Index of the correct answer in the `choices` array. |
| `correct_answer` | string | Original representation of the correct answer. |
| `correct_answer_plain` | string | Plain-text version of the correct answer. |

## Permissions and Ethical Considerations

The use of the dataset was carried out with the consent of doc. PaedDr. Monika Tomcsányiová, PhD., the national coordinator of the Slovak Bebras competition. This ensured that the data collection and usage complied with ethical and organizational considerations.

## Example Record

```json
{
"id": 70,
"serial_number": 5,
"title": "Semafory",
"question": "V Bobrove sú na každej križovatke semafory. Keď je na semafore zelená, autá môžú ísť rovno, doľava alebo doprava. Keď je červená, autá majú zastať a čakať na zelenú. Zelená a červená sa na každom semafore striedajú <b>každých 10 sekúnd</b>. Každý semafor má svoje počítadlo sekúnd, ktoré odpočítava od 10 do 0. Bobry vidia počítadlo a semafor zmení farbu vždy, keď dosiahne časovač 0. Vtedy sa časovač zmení na 10 a začne sa odpočítavanie pre druhú farbu.<br/> <image_1><br/> Bobrovi trvá 1 sekundu prejsť na ďalší štvorček, aj keď ide rovno, aj keď zatáča.<br/> Na mape sú znázornené ulice, ktoré spájajú bobrov dom a školu a tiež hodnoty počítadiel v okamihu, keď bobor vyrazí z domu (zo štvorčeka, na ktorom je auto).<br/> <image_2><br/> Bobor sa chce dostať do školy (modrý štvorček) tak, aby na žiadnom semafore <b>nemusel zastaviť</b> kvôli červenej.<br/> Ktorou trasou pôjde? <br/>",
"choices": [
{
"value": "0",
"label": "P→Q→S→R→T→škola"
},
{
"value": "1",
"label": "P→R→T→škola"
},
{
"value": "2",
"label": "P→Q→S→škola"
},
{
"value": "3",
"label": "P→R→S→škola"
}
],
"images": [
"http://demo.ibobor.sk/sutaz/images/20/2020-TW-03a_prejdi_stoj.png",
"http://demo.ibobor.sk/sutaz/images/20/2020-TW-03a.png"
],
"year": "2020/2021",
"category": "Senior",
"type": "image",
"correct_index": 3,
"correct_answer": "P→R→S→škola ",
"correct_answer_plain": "P→R→S→škola"
}