README.md

metadata

language:
  - kk
size_categories:
  - 1K-10K
task_categories:
  - visual-question-answering
  - multiple-choice
dataset_info:
  features:
    - name: index
      dtype: int64
    - name: question
      dtype: string
    - name: hint
      dtype: string
    - name: A
      dtype: string
    - name: B
      dtype: string
    - name: C
      dtype: string
    - name: D
      dtype: string
    - name: answer
      dtype: string
    - name: image
      dtype: image
  splits:
    - name: test
      num_examples: 4329
configs:
  - config_name: default
    data_files:
      - split: test
        path:
          - data/train-00000-of-00002.parquet
          - data/train-00001-of-00002.parquet

MMBench-Kazakh (Validation Split)

Summary

MMBench-Kazakh is a comprehensive translation of the MMBench (Validation Split) into the Kazakh language. MMBench is a popular multi-modality benchmark that evaluates models across various dimensions, including perception, reasoning, and logic, using a robust Multiple Choice Question (MCQ) format.

Data Fields

Field	Description
index	Original unique ID from the MMBench source.
question	The translated question in Kazakh.
hint	Optional context or hint for the question (translated).
A, B, C, D	Multiple-choice options translated into Kazakh.
image	Path to the associated image or visual context.

Example Instance

{
  "index": 42,
  "question": "Берілген суреттегі ең биік ғимарат қайсысы?",
  "hint": "Суретте қала панорамасы көрсетілген.",
  "A": "Көк ғимарат",
  "B": "Ортадағы мұнара",
  "C": "Оң жақтағы тұрғын үй",
  "D": "Сол жақтағы қонақ үй",
  "image": "images/42.png",
}

Funding

This dataset was developed as part of the project funded by the Ministry of Science and Higher Education of the Republic of Kazakhstan under Grant No. BR24993001, “Creation of a Large Language Model (LLM) to Support the Kazakh Language and Advance Technological Development.”