# CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities Reasoning Source: [CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities Reasoning]() Collection index: [llms.txt]() Data updated: 2026-03-10T05:00:12.119Z Authors: Yuxuan Liu, Weikai Xu, Kun Huang, Changyu Chen, Jiankun Zhao, Pengzhi Gao, Wei Liu, Jian Luan, Shuo Shang, Bo Du, Ji-Rong Wen, Rui Yan Published: 2026-02-27T16:19:45.000Z ## Abstract 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. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Mobile UI Reasoning on AMEX](): Mobile UI Reasoning; AMEX; Gmail Accuracy: 76.2. - [Action Prediction on AMEX (test)](): Action Prediction; AMEX (test); Action Accuracy: 72.61. - [Grounding on AMEX (test)](): Grounding; AMEX (test); Grounding Score: 79.19. - [Optical Character Recognition on AMEX (test)](): Optical Character Recognition; AMEX (test); OCR Accuracy: 79.67. - [Image Captioning on AMEX (test)](): Image Captioning; AMEX (test); Caption Score: 53.89. - [Action Prediction on AITZ (test)](): Action Prediction; AITZ (test); Scroll Accuracy: 52.07. ## Original sources and implementation [Original paper]() [PDF]()