---
license: apache-2.0
library_name: transformers
pipeline_tag: text-generation
base_model: Qwen/Qwen3-4B-Base
tags:
- qwen3
- reinforcement-learning
- agentic-reasoning
- verifiable-rewards
- vpr
- math
---

# VPR-Qwen3-4B-Base-Math-Mixed

This checkpoint is trained from Qwen3-4B-Base with mixed math and VPR game experience across Sokoban, Sudoku, and Minesweeper. VPR supplies action-level process rewards through task-grounded oracles and state-group rollout, while math training follows the mixed-training protocol described in the paper.

## Reported results

| Evaluation | Metric | Result |
|---|---|---:|
| General-reasoning OOD suite | Macro average | 52.75 |
| ALFWorld | SR | 16.12 ± 2.87 |
| WebShop | Score | 42.01 ± 2.14 |
| WebShop | SR | 1.33 ± 0.76 |

Values are reported under the VPR paper's evaluation protocol. **SR** is success rate. The general-reasoning value is the macro average across the reported OOD benchmarks. These OOD results are not a claim of exactly matched total rollout compute across training methods.

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "nics-efc/VPR-Qwen3-4B-Base-Math-Mixed"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

inputs = tokenizer("Solve the task step by step.", return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=1024)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

Use the [VPR codebase](https://github.com/thu-nics/VPR) for the exact prompts, environments, and evaluation entry points.

## Resources

- [Paper](https://arxiv.org/abs/2605.10325)
- [Code](https://github.com/thu-nics/VPR)
- [Project page](https://thu-nics.github.io/VPR/)

## Limitations

The checkpoint is shaped by the documented math distribution, task-grounded game oracles, prompts, and action formats. Performance and safety outside those settings have not been established.

## Citation

```bibtex
@misc{yuan2026verifiable,
  title         = {Verifiable Process Rewards for Agentic Reasoning},
  author        = {Huining Yuan and Zelai Xu and Huaijie Wang and Xiangmin Yi and Jiaxuan Gao and Xiao-Ping Zhang and Yu Wang and Chao Yu and Yi Wu},
  year          = {2026},
  eprint        = {2605.10325},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  url           = {https://arxiv.org/abs/2605.10325}
}
```