---
library_name: datasets
language:
- en
tags:
- multimodal
- reasoning
task_categories:
- visual-question-answering
size_categories:
- n<1K
configs:
- config_name: default
  data_files:
  - split: train
    path: dev120.parquet
---

# CoWorld-Reasoner-Dev

[![Paper](https://img.shields.io/badge/Paper-b31b1b?style=for-the-badge&logo=arxiv)](https://arxiv.org/abs/2610.21910)
[![Hugging Face Paper](https://img.shields.io/badge/Hugging%20Face-Paper-fcc21b?style=for-the-badge&logo=huggingface&logoColor=white)](https://huggingface.co/papers/2610.21910)
[![Code](https://img.shields.io/badge/GitHub-Code-blue?style=for-the-badge&logo=github)](https://github.com/ModalityDance/CoWorld-Reasoner)
[![Project Page](https://img.shields.io/badge/Project%20Page-GitHub-181717?style=for-the-badge&logo=github&logoColor=white)](https://modalitydance.github.io/CoWorld-Reasoner/)

## Overview

**CoWorld-Reasoner-Dev** contains the **120 development examples** used for harness co-evolution in CoWorld Reasoner, with **244 images**:

- **MMSI-Bench**: 30 examples
- **SpinBench**: 30 examples
- **CausalSpatial**: 30 examples
- **PerceptionBench**: 30 examples

These examples are excluded from the corresponding benchmark evaluations. The `train` split contains the development data, not the evaluation sets.

## Data Fields

Each example contains:

- `sample_id`: the example identifier.
- `benchmark`: the source benchmark.
- `task_type`: the task category.
- `answer_type`: the answer format.
- `question`: the question, including options when present.
- `options`: structured options when available.
- `images`: an ordered list of images embedded in the Parquet file.
- `image_paths`: the corresponding relative paths used by the code.
- `answer`: the reference answer.
- `score_spec`: the scoring method.

## Quick Start

### Load the dataset

```python
from datasets import load_dataset

ds = load_dataset("ModalityDance/CoWorld-Reasoner-Dev", split="train")
example = ds[0]

print(example["question"])
print(example["answer"])
print("Number of images:", len(example["images"]))
example["images"][0].show()
```

### Use with CoWorld Reasoner

After installing the [code repository](https://github.com/ModalityDance/CoWorld-Reasoner#installation), run:

```shell
python scripts/prepare/download.py
```

This saves questions and images to `data/dev120/` and reference answers to `scorer_bundle/data/dev120_evaluator_only.jsonl`.

## Acknowledgements

The examples are drawn from [MMSI-Bench](https://arxiv.org/abs/2505.23764v3), [SpinBench](https://arxiv.org/abs/2509.25390), [CausalSpatial](https://arxiv.org/abs/2601.13304), and [PerceptionBench](https://arxiv.org/abs/2607.24957). The source datasets retain their original licenses.

## Citation

Please cite our paper as:

```bibtex
@misc{cheng2026coworldreasoner,
  title  = {CoWorld Reasoner: Agentic Multimodal Reasoning with the Evolving Video World Model-Driven Harness},
  author = {Dongjie Cheng and Ruifeng Yuan and Ran Wei and Ning Song and Yongqi Li and Wenjie Li},
  year   = {2026},
  eprint = {2610.21910},
  archivePrefix = {arXiv},
  primaryClass = {cs.AI},
  url    = {https://arxiv.org/abs/2610.21910}
}
```