---
base_model:
- Qwen/Qwen2.5-1.5B-Instruct
language:
- en
- zh
license: cc-by-nc-nd-4.0
library_name: transformers
pipeline_tag: text-generation
tags:
- instruction-finetuning
- reasoning
- evaluation
inference: false
---

<h1 align="center">
🔍 xVerify-1.5B-I
</h1>

<p align="center">
  <div style="display: flex; justify-content: center; gap: 10px;">
    <a href="https://github.com/IAAR-Shanghai/xVerify">
      <img src="https://img.shields.io/badge/GitHub-Repository-blue?logo=github" alt="GitHub"/>
    </a>
    <a href="https://huggingface.co/IAAR-Shanghai/xVerify-1.5B-I">
      <img src="https://img.shields.io/badge/🤗%20Hugging%20Face-xVerify--1.5B--I-yellow" alt="Hugging Face"/>
    </a>
    <a href="https://huggingface.co/papers/2504.10481">
      <img src="https://img.shields.io/badge/Paper-HF%20Papers-red" alt="Paper"/>
    </a>
  </div>
</p>

xVerify is an evaluation tool fine-tuned from a pre-trained large language model, designed specifically for objective questions with a single correct answer. It was introduced in the paper [xVerify: Efficient Answer Verifier for Reasoning Model Evaluations](https://huggingface.co/papers/2504.10481).

The model accurately extracts the final answer from lengthy reasoning processes and efficiently identifies equivalence across different forms of expressions, helping to evaluate reasoning models that adopt slow-thinking strategies.

---

## ✨ Key Features

### 📊 Broad Applicability
Suitable for various objective question evaluation scenarios including math problems, multiple-choice questions, classification tasks, and short-answer questions.

### ⛓️ Handles Long Reasoning Chains
Effectively processes answers with extensive reasoning steps to extract the final answer, regardless of complexity.

### 🌐 Multilingual Support
Primarily handles Chinese and English responses while remaining compatible with other languages.

### 🔄 Powerful Equivalence Judgment
- **Basic Transformations**: Recognizes letter case changes and Greek letter conversions.
- **Mathematical Expressions**: Identifies equivalent expressions across formats like LaTeX, fractions, and scientific notation.
- **Semantic Equivalence**: Determines if natural language answers align with the correct reference.
- **Advanced Multiple-Choice**: Matches responses by content rather than just option identifiers.

---

## 📚 Citation

```bibtex
@article{xVerify,
      title={xVerify: Efficient Answer Verifier for Reasoning Model Evaluations}, 
      author={Ding Chen and Qingchen Yu and Pengyuan Wang and Wentao Zhang and Bo Tang and Feiyu Xiong and Xinchi Li and Minchuan Yang and Zhiyu Li},
      journal={arXiv preprint arXiv:2504.10481},
      year={2025},
}
```

---
*Authors: [Ding Chen](https://huggingface.co/Hush-cd), [Qingchen Yu](https://huggingface.co/Duguce), [Pengyuan Wang](https://huggingface.co/deflinhec), Wentao Zhang, Bo Tang, Feiyu Xiong, Xinchi Li, Minchuan Yang, Zhiyu Li.*