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

• URL: http://arxiv.org/abs/2509.21268v1
• Date: Thu, 25 Sep 2025 14:58:29 GMT
• Title: MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resources
• Authors: Sicong Leng, Jing Wang, Jiaxi Li, Hao Zhang, Zhiqiang Hu, Boqiang Zhang, Yuming Jiang, Hang Zhang, Xin Li, Lidong Bing, Deli Zhao, Wei Lu, Yu Rong, Aixin Sun, Shijian Lu,
• Abstract summary: Variance-Aware Sampling (VAS) is a data selection strategy guided by Variance Promotion Score (VPS)<n>We release large-scale, carefully curated resources containing 1.6M long CoT cold-start data and 15k RL QA pairs.<n> Experiments across mathematical reasoning benchmarks demonstrate the effectiveness of both the curated data and the proposed VAS.
• Score: 113.33902847941941
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large multimodal reasoning models have achieved rapid progress, but their advancement is constrained by two major limitations: the absence of open, large-scale, high-quality long chain-of-thought (CoT) data, and the instability of reinforcement learning (RL) algorithms in post-training. Group Relative Policy Optimization (GRPO), the standard framework for RL fine-tuning, is prone to gradient vanishing when reward variance is low, which weakens optimization signals and impairs convergence. This work makes three contributions: (1) We propose Variance-Aware Sampling (VAS), a data selection strategy guided by Variance Promotion Score (VPS) that combines outcome variance and trajectory diversity to promote reward variance and stabilize policy optimization. (2) We release large-scale, carefully curated resources containing ~1.6M long CoT cold-start data and ~15k RL QA pairs, designed to ensure quality, difficulty, and diversity, along with a fully reproducible end-to-end training codebase. (3) We open-source a family of 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. Comprehensive ablation studies and analyses provide further insight into the contributions of each component. In addition, we theoretically establish that reward variance lower-bounds the expected policy gradient magnitude, with VAS serving as a practical mechanism to realize this guarantee. Our code, data, and checkpoints are available at https://github.com/LengSicong/MMR1.

Related papers

• Controllable Exploration in Hybrid-Policy RLVR for Multi-Modal Reasoning [88.42566960813438]
  CalibRL is a hybrid-policy RLVR framework that supports controllable exploration with expert guidance.<n>CalibRL increases policy entropy in a guided manner and clarifies the target distribution.<n>Experiments across eight benchmarks, including both in-domain and out-of-domain settings, demonstrate consistent improvements.
  arXiv  Detail & Related papers  (2026-02-22T07:23:36Z)
• GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization [133.27496265096445]
  We show how to apply Group Relative Policy Optimization under multi-reward setting without examining its suitability.<n>We then introduce Group reward-Decoupled Normalization Policy Optimization (GDPO), a new policy optimization method to resolve these issues.<n>GDPO consistently outperforms GRPO, demonstrating its effectiveness and generalizability for multi-reward reinforcement learning optimization.
  arXiv  Detail & Related papers  (2026-01-08T18:59:24Z)
• Enhancing Agentic RL with Progressive Reward Shaping and Value-based Sampling Policy Optimization [13.475938754147625]
  Large Language Models (LLMs) empowered with Tool-Integrated Reasoning (TIR) can iteratively plan, call external tools, and integrate returned information to solve complex, long-horizon reasoning tasks.<n>Agentic Reinforcement Learning (Agentic RL) optimize such models over full tool-interaction trajectories.<n>Two key challenges hinder effectiveness: (1) Sparse, non-instructive rewards, such as binary 0-1 verifiable signals, provide limited guidance for intermediate steps and slow convergence.<n>We propose two complementary techniques: Progressive Reward Shaping (PRS) and Value-based Sampling Policy Optimization (VSPO).
  arXiv  Detail & Related papers  (2025-12-08T11:59:25Z)
• VADE: Variance-Aware Dynamic Sampling via Online Sample-Level Difficulty Estimation for Multimodal RL [38.782188833641676]
  Group-based policy optimization methods like GRPO and GSPO have become standard for training multimodal models.<n>They suffer from a critical emphgradient vanishing problem when all responses within a group receive identical rewards.<n>We propose textbfVADE, a sampling framework via online sample-level difficulty textbfEstimation.
  arXiv  Detail & Related papers  (2025-11-24T08:59:54Z)
• CoT-Saliency: Unified Chain-of-Thought Reasoning for Heterogeneous Saliency Tasks [96.64597365827046]
  We present the first unified framework that jointly handles three operationally heterogeneous saliency tasks.<n>We introduce a Chain-of-Thought (CoT) reasoning process in a Vision-Language Model (VLM) to bridge task heterogeneity.<n>We show our model matches or outperforms specialized SOTA methods and strong closed-source VLMs across all tasks.
  arXiv  Detail & Related papers  (2025-11-01T04:37:01Z)
• Reinforcement Learning for Multi-Objective Multi-Echelon Supply Chain Optimisation [3.1194372040101928]
  The model is evaluated using a multi-objective reinforcement learning (RL) method, benchmarked against an originally single-objective RL algorithm modified with weighted sum.<n>We conduct experiments on varying network complexities, mimicking typical real-world challenges using a customisable simulator.<n>The model determines production and delivery quantities across supply chain routes to achieve near-optimal trade-offs between competing objectives.
  arXiv  Detail & Related papers  (2025-07-26T04:30:11Z)
• Improving Multimodal Learning via Imbalanced Learning [6.93254775445168]
  Multimodal learning often encounters the under-optimized problem and may perform worse than unimodal learning.<n>This paper argues that balanced learning is not the optimal setting for multimodal learning.<n>We propose the Asymmetric Representation Learning(ARL) strategy to assist multimodal learning via imbalanced optimization.
  arXiv  Detail & Related papers  (2025-07-14T12:14:57Z)
• Reinforcing Video Reasoning with Focused Thinking [65.85683941058916]
  We propose TW-GRPO, a novel framework that enhances visual reasoning with focused thinking and dense reward granularity.<n>Specifically, we employ a token weighting mechanism that prioritizes tokens with high informational density.<n>We also reformulate RL training by shifting from single-choice to multi-choice QA tasks.
  arXiv  Detail & Related papers  (2025-05-30T15:42:19Z)
• Flow-GRPO: Training Flow Matching Models via Online RL [75.70017261794422]
  We propose Flow-GRPO, the first method integrating online reinforcement learning (RL) into flow matching models.<n>Our approach uses two key strategies: (1) an ODE-to-SDE conversion that transforms a deterministic Ordinary Equation (ODE) into an equivalent Differential Equation (SDE) that matches the original model's marginal distribution at all timesteps; and (2) a Denoising Reduction strategy that reduces training denoising steps while retaining the original inference timestep number.
  arXiv  Detail & Related papers  (2025-05-08T17:58:45Z)
• A Distribution-Aware Flow-Matching for Generating Unstructured Data for Few-Shot Reinforcement Learning [1.0709300917082865]
  We introduce a distribution-aware flow matching approach to generate synthetic unstructured data for few-shot reinforcement learning.<n>Our approach addresses key challenges in traditional model-based RL, such as overfitting and data correlation.<n>Results demonstrate that our method achieves stable convergence in terms of maximum Q-value while enhancing frame rates by 30% in the initial timestamps.
  arXiv  Detail & Related papers  (2024-09-21T15:50:59Z)
• Preference-Based Multi-Agent Reinforcement Learning: Data Coverage and Algorithmic Techniques [65.55451717632317]
  We study Preference-Based Multi-Agent Reinforcement Learning (PbMARL)<n>We identify the Nash equilibrium from a preference-only offline dataset in general-sum games.<n>Our findings underscore the multifaceted approach required for PbMARL.
  arXiv  Detail & Related papers  (2024-09-01T13:14:41Z)
• Multifidelity Reinforcement Learning with Control Variates [3.2895195535353317]
  In many computational science and engineering applications, the output of a system of interest corresponding to a given input can be queried at different levels of fidelity with different costs. We study the reinforcement learning problem in the presence of multiple environments with different levels of fidelity for a given control task. A multifidelity estimator that exploits the cross-correlations between the low- and high-fidelity returns is proposed to reduce the variance in the estimation of the state-action value function.
  arXiv  Detail & Related papers  (2022-06-10T15:01:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.