---
license: cc
task_categories:
- multiple-choice
- visual-question-answering
- reinforcement-learning
language:
- en
tags:
- MCQA
- Multimodal
- VQA
- MCQ
- RL
- RLVR
size_categories:
- 100K<n<1M
configs:
- config_name: default
  data_files:
  - split: validation
    path: data/validation-*
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
dataset_info:
  features:
  - name: photoid
    dtype: int64
  - name: downloadurl
    dtype: string
  - name: question
    dtype: string
  - name: options
    dtype: string
  - name: answer
    dtype: string
  - name: image
    dtype: 'null'
  - name: resolution
    dtype: string
  splits:
  - name: validation
    num_bytes: 621106
    num_examples: 3500
  - name: train
    num_bytes: 16465030
    num_examples: 93000
  - name: test
    num_bytes: 619592
    num_examples: 3500
  download_size: 4359647
  dataset_size: 17705728
---

http://arxiv.org/abs/2509.25941