MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources

Introduction

This repository introduces the MMR1 project, focusing on enhancing large multimodal reasoning models. While rapid progress has been made, advancements are constrained by two major limitations:

1. The absence of open, large-scale, high-quality long chain-of-thought (CoT) data.
2. The instability of reinforcement learning (RL) algorithms in post-training, where standard Group Relative Policy Optimization (GRPO) is prone to gradient vanishing under low reward variance, weakening optimization signals and impairing convergence.

MMR1 addresses these challenges through several key contributions:

• Variance-Aware Sampling (VAS): A novel data selection strategy guided by the Variance Promotion Score (VPS). VAS combines outcome variance and trajectory diversity to promote reward variance, stabilize policy optimization, and improve convergence.
• Large-scale curated resources: The project releases carefully curated datasets containing ~1.6M long CoT cold-start data and ~15k RL QA pairs, designed to ensure quality, difficulty, and diversity.
• Open-source codebase & models: A fully reproducible end-to-end training pipeline and a family of open-sourced multimodal reasoning models in multiple scales, establishing standardized baselines for the community.

Experiments across mathematical reasoning benchmarks demonstrate the effectiveness of both the curated data and the proposed VAS.

Open Resources

We release the following datasets for the community:

• MMR1-SFT (~1.6M): A supervised fine-tuning dataset with 1.6 million long Chain-of-Thought (CoT) cold-start trajectories. These trajectories were generated using Gemini2.5 Pro/Flash and verified with GPT-4o to ensure high quality and accuracy.
• MMR1-RL (15k): An RL dataset consisting of 15,000 question-answer pairs, curated with GPT-4o. This dataset is designed to provide robust reward signals for policy optimization.

The dataset spans diverse domains—including mathematics, science, charts/figures, document tables, and general understanding—covering ~1.6M math samples and an additional ~37K samples across other domains. It integrates existing public resources (e.g., MathVerse, ScienceQA, ChartQA, DocVQA, GQA) together with newly curated and self-collected data, ensuring quality, difficulty, and diversity. This collection establishes one of the most comprehensive open resources for multimodal reasoning models.

We hope these resources can serve as a benchmark for the community and facilitate the research of multimodal reasoning.

Citation

If you find MMR1 useful for your research and applications, please cite using this BibTeX:

@misc{leng2025mmr1,
  title={MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources}, 
  author={Sicong Leng and Jing Wang and Jiaxi Li and Hao Zhang and Zhiqiang Hu and Boqiang Zhang and Yuming Jiang and Hang Zhang and Xin Li and Lidong Bing and Deli Zhao and Wei Lu and Yu Rong and Aixin Sun and Shijian Lu},
  year={2025},
  eprint={2509.21268},
  archivePrefix={arXiv},
  primaryClass={cs.CV},
  url={https://arxiv.org/abs/2509.21268}, 
}

License

This project is released under the Apache 2.0 license as found in the LICENSE file in the GitHub repository. The service is a research preview intended for non-commercial use ONLY, subject to the model Licenses of Qwen, Terms of Use of the data generated by OpenAI and Gemini, and Privacy Practices of ShareGPT. Please get in touch with the authors if you find any potential violations.