CVOct 22, 2025

DaMo: Data Mixing Optimizer in Fine-tuning Multimodal LLMs for Mobile Phone Agents

arXiv:2510.19336v11 citationsh-index: 6Has Code
Originality Incremental advance
AI Analysis

This addresses the challenge of multitask learning for mobile phone agents, offering an incremental improvement in data optimization for fine-tuning MLLMs.

The paper tackles the problem of optimizing data mixtures for fine-tuning multimodal large language models (MLLMs) in mobile phone agents, proposing DaMo, a trainable network that predicts optimal data compositions, resulting in a 3.38% performance improvement on a new benchmark and up to 12.47% gains on specific tasks.

Mobile Phone Agents (MPAs) have emerged as a promising research direction due to their broad applicability across diverse scenarios. While Multimodal Large Language Models (MLLMs) serve as the foundation for MPAs, their effectiveness in handling multiple mobile phone tasks simultaneously remains limited. Although multitask supervised fine-tuning (SFT) is widely adopted for multitask learning, existing approaches struggle to determine optimal training data compositions for peak performance. To address this challenge, we propose DaMo (Data Mixture Optimizer) - a novel solution employing a trainable network that predicts optimal data mixtures by forecasting downstream task performance for any given dataset ratio. To support comprehensive evaluation, we introduce PhoneAgentBench, the first specialized benchmark to evaluate MLLMs on multimodal mobile phone tasks, comprising 1235 QA pairs spanning diverse real-world industrial mobile application scenarios. Demonstrating strong predictive capability (R^2=0.81) in small-scale pilot experiments, DaMo efficiently extrapolates optimal data mixing configurations. Our results show DaMo achieves a 3.38% performance improvement on PhoneAgentBench compared to alternative methods. Furthermore, extensive experiments across established benchmarks including BFCL-v3, MME-Reasoning, MME-Perception, and OCRBench reveal DaMo's superior generalization, outperforming other approaches by 2.57% in terms of average score. When used solely for MLLM optimization on the BFCL-v3 task, DaMo improves the metrics by 12.47% than other methods. Notably, DaMo maintains robust scalability, preserving its effectiveness when applied to other model architectures. The code and dataset are available at https://github.com/OPPO-Mente-Lab/DaMo.git

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes