CoME：赋能移动专家的 informative hybrid-capabilities 推理
CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities Reasoning
AI总结：
CoME 通过四种专家架构和渐进训练策略，实现移动代理混合能力推理的解耦增强与平衡优化。
摘要：
移动代理可以自主执行用户指令，这需要混合能力推理，包括屏幕摘要、子任务规划、行动决策和行动功能。然而，现有代理难以实现这些能力的解耦增强和平衡整合。为解决这些问题，我们提出了 Channel-of-Mobile-Experts（CoME），一种由四个不同专家组成的新型代理架构，每个专家对应特定的推理阶段。CoME通过输出导向激活激活相应的专家以生成输出标记。为了使CoME具备混合能力推理能力，我们引入了一种渐进训练策略：Expert-FT 用于解耦和增强不同专家的能力；Router-FT 用于对齐专家激活与不同的推理阶段；CoT-FT 促进多个能力之间的无缝协作和平衡优化。为了缓解混合能力推理中的误差传播，我们提出了 InfoGain-Driven DPO（Info-DPO），它使用信息增益来评估每个中间步骤的贡献，从而引导 CoME 向更信息丰富的推理发展。全面实验表明，CoME 在 AITZ 和 AMEX 数据集上均优于密集移动代理和 MoE 方法。
英文摘要：
Mobile Agents can autonomously execute user instructions, which requires hybrid-capabilities reasoning, including screen summary, subtask planning, action decision and action function. However, existing agents struggle to achieve both decoupled enhancement and balanced integration of these capabilities. To address these challenges, we propose Channel-of-Mobile-Experts (CoME), a novel agent architecture consisting of four distinct experts, each aligned with a specific reasoning stage, CoME activates the corresponding expert to generate output tokens in each reasoning stage via output-oriented activation. To empower CoME with hybrid-capabilities reasoning, we introduce a progressive training strategy: Expert-FT enables decoupling and enhancement of different experts' capability; Router-FT aligns expert activation with the different reasoning stage; CoT-FT facilitates seamless collaboration and balanced optimization across multiple capabilities. To mitigate error propagation in hybrid-capabilities reasoning, we propose InfoGain-Driven DPO (Info-DPO), which uses information gain to evaluate the contribution of each intermediate step, thereby guiding CoME toward more informative reasoning. Comprehensive experiments show that CoME outperforms dense mobile agents and MoE methods on both AITZ and AMEX datasets.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
