---
license: apache-2.0
base_model:
- Qwen/Qwen3-VL-8B-Instruct
---
# Model Card for [PEARL-8B](https://github.com/MiliLab/PEARL)-Based on [Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct)

<!-- Provide a quick summary of what the model is/does. -->

Perceptual-Evidence Anchored Reinforced Learning for Multimodal Reasoning.
arxiv.org/abs/2511.18437
## Model Details

### Model Description
This is a multimodal reasoning model. 
<!-- Provide a longer summary of what this model is. -->



- **Developed by:** [Chi Zhang~1909zczc@gmail.com]
- **Finetuned from model [optional]:** [Qwen3-VL-8B-Instruct]

### Model Sources [optional]

<!-- Provide the basic links for the model. -->

- **Repository:** [[PEARL](https://github.com/MiliLab/PEARL)]
- **Paper:** [More Information Needed]

## Uses
[VLMEvalkit](https://github.com/open-compass/VLMEvalKit)

## Training Details
[EasyR1](https://github.com/hiyouga/EasyR1)
### Training Data
[ViRL39k](https://huggingface.co/datasets/TIGER-Lab/ViRL39K)


## Citation [optional]

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->