---
datasets:
- Fish-03/RuleMaze
base_model:
- Qwen/Qwen2.5-VL-3B-Instruct
library_name: peft
tags:
- multimodal
- vision-language
- visual-planning
- spatial-planning
- rule-following
- qwen2.5-vl
- lora
pipeline_tag: image-text-to-text
---

# RuleMaze

RuleMaze is a benchmark and training framework for **rule-compliant visual spatial planning** with Multimodal Large Language Models (MLLMs).

Given a visual maze and a set of natural-language rules, the model is required to understand the environment, follow the rules, and generate a valid multi-step trajectory.

This repository provides LoRA adapters fine-tuned from **Qwen2.5-VL-3B-Instruct** on the RuleMaze training data.

## Checkpoints

Two scene types are provided:

* `RuleMaze/regular/checkpoint`
* `RuleMaze/quest/checkpoint`

The checkpoints are PEFT/LoRA adapters and should be loaded together with the base model:

```text
Qwen/Qwen2.5-VL-3B-Instruct
```

## Resources

* **Code:** https://github.com/oceanflowlab/RuleMaze
* **Dataset:** https://huggingface.co/datasets/Fish-03/RuleMaze
* **Base Model:** https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct

## Training and Evaluation

The models are trained using the RuleMaze DMP training pipeline with LLaMA-Factory.

RuleMaze evaluates visual planning under both **seen-rule** and **unseen-rule** settings and different rule difficulties.

For training and evaluation details, please refer to the official code repository.

## Intended Use

The models are intended for research on:

* multimodal reasoning
* visual spatial planning
* rule following
* compositional generalization

## Citation

If you find RuleMaze useful, please cite:

```bibtex
@misc{rulemaze,
  title  = {Rule-Compliant Visual Spatial Planning for Multimodal Large Language Models},
  author = {Yu Chen, Ting Lei, Yaoyi Li, Jia Cai, Zhecen Wu and Yang Liu},
  year   = {2026},
  note   = {Code and dataset release}
}
```