CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities Reasoning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917317421563904 |
|---|---|
| author | Liu, Yuxuan Xu, Weikai Huang, Kun Chen, Changyu Zhao, Jiankun Gao, Pengzhi Liu, Wei Luan, Jian Shang, Shuo Du, Bo Wen, Ji-Rong Yan, Rui |
| author_facet | Liu, Yuxuan Xu, Weikai Huang, Kun Chen, Changyu Zhao, Jiankun Gao, Pengzhi Liu, Wei Luan, Jian Shang, Shuo Du, Bo Wen, Ji-Rong Yan, Rui |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_24142 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities Reasoning Liu, Yuxuan Xu, Weikai Huang, Kun Chen, Changyu Zhao, Jiankun Gao, Pengzhi Liu, Wei Luan, Jian Shang, Shuo Du, Bo Wen, Ji-Rong Yan, Rui Computation and Language Artificial Intelligence 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. |
| title | CoME: Empowering Channel-of-Mobile-Experts with Informative Hybrid-Capabilities Reasoning |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2602.24142 |