UniReason-Qwen3-14B-RL/README.md

---
base_model: qwen3-14b
datasets:
- math
- reasoning
language: en
license: apache-2.0
pipeline_tag: text-generation
tags:
- text-generation
- math-reasoning
- transferability
- RL-GRPO
- research-paper
- qwen
arxiv: 2507.00432
library_name: transformers
---

# UniReason-Qwen3-14B-RL

This model is associated with the research paper:
**"Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning"**

📄 **Paper**: [2507.00432](https://arxiv.org/abs/2507.00432)
💻 **Code**: [https://github.com/ReasoningTransfer/Transferability-of-LLM-Reasoning](https://github.com/ReasoningTransfer/Transferability-of-LLM-Reasoning)

## Abstract

Math reasoning has become the poster child of progress in large language models (LLMs), with new models rapidly surpassing human-level performance on benchmarks like MATH and AIME. But as math leaderboards improve week by week, it is worth asking: do these gains reflect broader problem-solving ability or just narrow overfitting?

## Model Description

This model is a **RL-GRPO**-tuned version of qwen3-14b focused on **math-reasoning** capabilities.
The model was developed as part of research investigating the transferability of mathematical reasoning skills to general language tasks.

### Key Research Questions Addressed:
- Does math reasoning training improve general LLM capabilities?
- How do different training methods (RL vs SFT) affect transferability?
- What is the trade-off between specialized math performance and general capabilities?

## Model Details

- **Base Model**: qwen3-14b
- **Training Method**: RL-GRPO
- **Primary Focus**: math-reasoning
- **Training Data**: Math-specific datasets
- **Architecture**: Transformer-based language model
- **Parameters**: 14B

## Training Details

### Training Method: RL-GRPO
Custom training methodology - see paper for details.

### Datasets Used
- Mathematical reasoning datasets
- See paper for complete dataset list

## Performance

### Math Reasoning Benchmarks
- **MATH**: See paper
- **AIME**: See paper

### General Capabilities
- **General QA**: See paper
- **Code Generation**: See paper
- **Instruction Following**: See paper

*For detailed performance metrics, please refer to the paper.*

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# Load model and tokenizer
model_name = "ReasoningTransferability/UniReason-Qwen3-14B-RL"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto"
)

# Example: Math reasoning
math_prompt = "Solve this step by step: What is the derivative of x^3 + 2x^2 - 5x + 1?"
inputs = tokenizer(math_prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=512, temperature=0.7)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

# Example: General reasoning
general_prompt = "Explain the concept of supply and demand in economics."
inputs = tokenizer(general_prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=512, temperature=0.7)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)
```

## Limitations and Biases

- **Specialization Trade-offs**: As explored in the paper, models optimized for math reasoning may show reduced performance on general tasks
- **Training Method Dependencies**: Performance characteristics vary significantly between RL and SFT training approaches
- **Domain Transfer**: The extent of capability transfer from math to other domains is limited
- **Computational Requirements**: Model requires significant computational resources for inference

## Research Findings

Key findings from the associated paper:
1. **RL vs SFT**: RL-tuned models show better transfer to general domains compared to SFT-tuned models
2. **Capability Trade-offs**: Most math-specialized models fail to transfer gains to other domains
3. **Forgetting**: SFT-tuned models often forget general capabilities during math-focused training

## Ethical Considerations

- This model is intended for research purposes
- Users should be aware of potential biases in mathematical and general reasoning
- The model should not be used for making critical decisions without human oversight
- Consider the environmental impact of large model inference

## Citation

If you use this model in your research, please cite both the model and the associated paper:

```bibtex
@misc{huan2025doesmathreasoningimprove,
      title={Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning}, 
      author={Maggie Huan and Yuetai Li and Tuney Zheng and Xiaoyu Xu and Seungone Kim and Minxin Du and Radha Poovendran and Graham Neubig and Xiang Yue},
      year={2025},
      eprint={2507.00432},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2507.00432}, 
}
```

## Contact

For questions about this model or the associated research, please:
- Open an issue in this repository
- Contact the paper authors
- Reference the original paper: https://arxiv.org/abs/2507.00432

## Acknowledgments

This work builds upon the research presented in "Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning" and uses the qwen3-14b architecture as its foundation.

---

*Model uploaded on 2025-07-03*