MoDoMoDo: Multi-Domain Data Mixtures for Multimodal LLM Reinforcement Learning

• URL: http://arxiv.org/abs/2505.24871v2
• Date: Thu, 05 Jun 2025 05:13:46 GMT
• Title: MoDoMoDo: Multi-Domain Data Mixtures for Multimodal LLM Reinforcement Learning
• Authors: Yiqing Liang, Jielin Qiu, Wenhao Ding, Zuxin Liu, James Tompkin, Mengdi Xu, Mengzhou Xia, Zhengzhong Tu, Laixi Shi, Jiacheng Zhu,
• Abstract summary: Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as a powerful paradigm for post-training large language models (LLMs)<n>Applying RLVR to Multimodal LLMs (MLLMs) presents significant opportunities but is complicated by the broader, heterogeneous nature of vision-language tasks.<n>We introduce a systematic post-training framework for Multimodal LLM RLVR, featuring a rigorous data mixture problem formulation and benchmark implementation.
• Score: 37.71233459623324
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as a powerful paradigm for post-training large language models (LLMs), achieving state-of-the-art performance on tasks with structured, verifiable answers. Applying RLVR to Multimodal LLMs (MLLMs) presents significant opportunities but is complicated by the broader, heterogeneous nature of vision-language tasks that demand nuanced visual, logical, and spatial capabilities. As such, training MLLMs using RLVR on multiple datasets could be beneficial but creates challenges with conflicting objectives from interaction among diverse datasets, highlighting the need for optimal dataset mixture strategies to improve generalization and reasoning. We introduce a systematic post-training framework for Multimodal LLM RLVR, featuring a rigorous data mixture problem formulation and benchmark implementation. Specifically, (1) We developed a multimodal RLVR framework for multi-dataset post-training by curating a dataset that contains different verifiable vision-language problems and enabling multi-domain online RL learning with different verifiable rewards; (2) We proposed a data mixture strategy that learns to predict the RL fine-tuning outcome from the data mixture distribution, and consequently optimizes the best mixture. Comprehensive experiments showcase that multi-domain RLVR training, when combined with mixture prediction strategies, can significantly boost MLLM general reasoning capacities. Our best mixture improves the post-trained model's accuracy on out-of-distribution benchmarks by an average of 5.24% compared to the same model post-trained with uniform data mixture, and by a total of 20.74% compared to the pre-finetuning baseline.

Related papers

• IDEAL: Data Equilibrium Adaptation for Multi-Capability Language Model Alignment [29.703775936837012]
  Large Language Models (LLMs) have achieved impressive performance through Supervised Fine-tuning (SFT) on diverse instructional datasets.<n>When training on multiple capabilities simultaneously, the mixture training dataset, governed by volumes of data from different domains, is a critical factor that directly impacts the final model's performance.<n>We introduce an innovative data equilibrium framework designed to effectively optimize volumes of data from different domains within mixture SFT datasets.
  arXiv  Detail & Related papers  (2025-05-19T06:42:44Z)
• EMORL: Ensemble Multi-Objective Reinforcement Learning for Efficient and Flexible LLM Fine-Tuning [6.675088737484839]
  We introduce an Ensemble Multi-Objective RL (EMORL) framework that fine-tunes multiple models with individual objectives to improve efficiency and flexibility.<n>Our method is the first to aggregate the hidden states of individual models, incorporating contextual information from multiple objectives.<n>We demonstrate the advantages of EMORL against existing baselines in experiments on the PAIR and Psych8k datasets.
  arXiv  Detail & Related papers  (2025-05-05T11:30:46Z)
• Diversity as a Reward: Fine-Tuning LLMs on a Mixture of Domain-Undetermined Data [34.6322241916799]
  Fine-tuning large language models (LLMs) using diverse datasets is crucial for enhancing their overall performance across various domains.<n>We propose a new method that gives the LLM a dual identity: an output model to cognitively probe and select data based on diversity reward, as well as an input model to be tuned with the selected data.
  arXiv  Detail & Related papers  (2025-02-05T17:21:01Z)
• GME: Improving Universal Multimodal Retrieval by Multimodal LLMs [43.457928045291915]
  Universal Multimodal Retrieval (UMR) aims to enable search across various modalities using a unified model.<n>Previous work has attempted to adopt multimodal large language models (MLLMs) to realize UMR using only text data.
  arXiv  Detail & Related papers  (2024-12-22T04:40:24Z)
• FedMLLM: Federated Fine-tuning MLLM on Multimodal Heterogeneity Data [56.08867996209236]
  Fine-tuning Multimodal Large Language Models (MLLMs) with Federated Learning (FL) allows for expanding the training data scope by including private data sources.<n>We introduce a benchmark to evaluate the performance of federated fine-tuning of MLLMs across various multimodal heterogeneous scenarios.<n>We develop a general FedMLLM framework that integrates classic FL methods alongside two modality-agnostic strategies.
  arXiv  Detail & Related papers  (2024-11-22T04:09:23Z)
• Star-Agents: Automatic Data Optimization with LLM Agents for Instruction Tuning [71.2981957820888]
  We propose a novel Star-Agents framework, which automates the enhancement of data quality across datasets. The framework initially generates diverse instruction data with multiple LLM agents through a bespoke sampling method. The generated data undergo a rigorous evaluation using a dual-model method that assesses both difficulty and quality.
  arXiv  Detail & Related papers  (2024-11-21T02:30:53Z)
• NVLM: Open Frontier-Class Multimodal LLMs [64.00053046838225]
  We introduce NVLM 1.0, a family of frontier-class multimodal large language models (LLMs) that achieve state-of-the-art results on vision-language tasks. We propose a novel architecture that enhances both training efficiency and multimodal reasoning capabilities. We develop production-grade multimodality for the NVLM-1.0 models, enabling them to excel in vision-language tasks.
  arXiv  Detail & Related papers  (2024-09-17T17:59:06Z)
• A Framework for Fine-Tuning LLMs using Heterogeneous Feedback [69.51729152929413]
  We present a framework for fine-tuning large language models (LLMs) using heterogeneous feedback. First, we combine the heterogeneous feedback data into a single supervision format, compatible with methods like SFT and RLHF. Next, given this unified feedback dataset, we extract a high-quality and diverse subset to obtain performance increases.
  arXiv  Detail & Related papers  (2024-08-05T23:20:32Z)
• BiMix: A Bivariate Data Mixing Law for Language Model Pretraining [47.77701041534746]
  The impact of pretraining data composition on model performance remains poorly understood.<n>$textbfBiMix$ provides a systematic framework for understanding and optimizing data mixtures.<n>Our work contributes both theoretical insights into data mixing dynamics and practical tools for enhancing LLM training efficiency.
  arXiv  Detail & Related papers  (2024-05-23T09:44:02Z)
• How to Train Data-Efficient LLMs [56.41105687693619]
  We study data-efficient approaches for pre-training language models (LLMs) We find that Ask-LLM and Density sampling are the best methods in their respective categories. In our comparison of 19 samplers, involving hundreds of evaluation tasks and pre-training runs, we find that Ask-LLM and Density are the best methods in their respective categories.
  arXiv  Detail & Related papers  (2024-02-15T02:27:57Z)

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.