---
title: Vietnamese-Entrance-Exam
canonical_url: "https://www.modelscope.cn/datasets/Intelligent-Internet/Vietnamese-Entrance-Exam"
md_url: "https://www.modelscope.cn/datasets/Intelligent-Internet/Vietnamese-Entrance-Exam.md"
repository: Intelligent-Internet/Vietnamese-Entrance-Exam
last_updated: 2025-03-31
license: "Apache License 2.0"
storage_size: "52 MB"
downloads: 930
stars: 0
---

# Vietnamese-Entrance-Exam

> Vietnamese-Entrance-Exam - Intelligent-Internet 在 ModelScope 开源的数据集。Vietnamese Entrance Exam Dataset

Intelligent-Internet/Vietnamese-Entrance-Exam 是 ModelScope 魔搭社区上的数据集，存储大小 52 MB，采用 Apache License 2.0 许可。

- **Repository**: Intelligent-Internet/Vietnamese-Entrance-Exam
- **License**: Apache License 2.0
- **Storage size**: 52 MB
- **Downloads**: 930
- **Stars**: 0
- **Last updated**: 2025-03-31

Source: https://www.modelscope.cn/datasets/Intelligent-Internet/Vietnamese-Entrance-Exam

---

## Vietnamese Entrance Exam Dataset

The Vietnamese Entrance Exam dataset is a collection of 432 problems derived from Vietnamese University entrance examinations. The dataset aims to provide a novel benchmark for testing reasoning capabilities of language models in several low resource domains specifically designed to minimize potential data contamination from pre-training or post-training exposure.
|Domain | Count| 
|--------|--------|
|Physics| 95|
|Chemistry| 94|
|Math | 243|

### Data Collection Process
1. Source Collection
* Problems gathered from high-quality TeX sources.
* Additional problems extracted via OCR using Gemini-2.0-Flash.
2. Translation
* Original Vietnamese problems translated to English.
3. Reformulation Process
* Multiple-choice questions converted to direct numerical answer format.
* Transformation process inspired by BigMath methodology.
* Each question reformulated using predefined criteria.
4. Validation
* LLM judge evaluation of reformulated questions.
* Verification of transformation validity.
* Final rewriting based on judge's criteria.

### Benchmark

| Model                          | Chemistry benchmark | Physic benchmark | Math benchmark |
|-------------------------------|---------------------|------------------|----------------|
| O1                            | 21.27               | 52.63            | 50.2           |
| O1-mini                       | 22.34               | 56.84            | 65.02          |
| O3-mini                       | 18.08               | 60.00            | 72.48          |
| DeepSeek-R1                   | **30.85**              | **74.73**            | 80.24          |
| DeepSeek-R1-Distill-Qwen-32B | 19.14               | 57.89            | 72.43          |
|Qwen/QwQ-32B| 26.59 |73.68|**81.89**|

### Citation

If you find our work useful, please cite our technical report:

```bib
@misc{2025iithought,
      title={II-Thought : A Large-Scale, High-Quality Reasoning Dataset},
      author={Intelligent Internet}
      year={2025},
}
```