base_model:
- Qwen/Qwen3.5-4B
license: mit
pipeline_tag: text-generation
Distillation-Learner-Qwen3.5-4B
🌐 Project Page | 💻 Code | 📄 Paper
We introduce RecursiveMAS, a multi-agent framework that scales agent collaboration through latent-space recursion. RecursiveMAS treats a multi-agent system as a unified recursive computation, where heterogeneous agents iteratively exchange, refine, and evolve their latent states across recursion rounds. In the Distillation-Style setting, the Learner Agent receives guidance from the Expert Agent and performs task solving through recursive latent-space collaboration.
Model Details
| Item | Description |
|---|---|
| Model | Distillation-Learner-Qwen3.5-4B |
| Collaboration Style | Distillation-Style |
| Agent Role | Learner Agent |
| Base Model | Qwen3.5-4B |
⚠️ Note: This checkpoint is a role-specific agent in RecursiveMAS, rather than a standalone model intended for plain-text generation.
Usage
For detailed environment setup and full training/inference instructions, please refer to the GitHub repository. You can load the multi-agent system containing this agent using the following snippet:
from system_loader import load_mas_system
mas = load_mas_system(
style="distillation",
device="cuda",
trust_remote_code=True,
)
expert = mas.agents["expert"].model
learner = mas.agents["learner"].model
Model Collections for RecursiveMAS
| Style | Model Collection |
|---|---|
| Sequential-Style | 🤗 HuggingFace |
| Mixture-Style | 🤗 HuggingFace |
| Distillation-Style | 🤗 HuggingFace |
| Deliberation-Style | 🤗 HuggingFace |
Experiment Results
Citation
@misc{recursivemas,
title={Recursive Multi-Agent Systems},
author={Xiyuan Yang and Jiaru Zou and Rui Pan and Ruizhong Qiu and Pan Lu and Shizhe Diao and Jindong Jiang and Hanghang Tong and Tong Zhang and Markus J. Buehler and Jingrui He and James Zou},
year={2026},
eprint={2604.25917},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2604.25917},
}