Newest papers

Tandem: Serving Hybrid Linear-Attention GUI Agents on Mobile NPUs

Zhuomin Chen, Jingchao Ni, Xu Zheng, Yong Chen, Mo Sha, Wei Cheng, Dongsheng Luo · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26660 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Agen latent GUI dari agen screenshot mencatat tugas-tugas dalam aplikasi seluler dengan berulang kali membaca layar dan meminta model untuk tindakan berikutnya. Unit pemrosesan saraf seluler (NPU) memungkinkan eksekusi lokal, mempertahankan konten layar di telepon, tetapi latensi model menumpuk dengan setiap tindakan. Model perhatian linier hibrida mengurangi persyaratan memori, namun menggunakan kembali keadaan mereka yang berulang membutuhkan titik check pada posisi input yang tepat. Pada permintaan agen mempertahankan instruksi tetapi menulis ulang riwayat baru-baru ini, sehingga kebijakan titik checkpoint yang dirancang untuk konfigurasi obrolan dapat mengevaluasi prefix yang dapat digunakan kembali. Mempertahankan status rollback teks untuk decoding spekulatif juga dapat memindahkan proses prompt dari unit matriks NPU, menambahkan overhead yang melebihi tabungan dekoding pada output aksi pendek.

Machine translation (Meta M2M100); the English original follows.

Abstract

Smartphone GUI agents complete tasks in mobile apps by repeatedly reading the screen and asking a model for the next action. Mobile neural processing units (NPUs) enable local execution, keeping screen contents on the phone, but model latency accumulates with every action. Hybrid linear attention models reduce memory requirements, yet reusing their recurrent state requires checkpoints at the appropriate input positions. Agent requests retain instructions but rewrite recent history, so checkpoint policies designed for chat continuation can miss reusable prefixes. Retaining rollback states for speculative decoding can also move prompt processing off the NPU's matrix unit, adding overhead that outweighs decoding savings on short action outputs. A loaded model can further hold the NPU awake between tasks. We present Tandem, a runtime for hybrid GUI agents on mobile NPUs. Tandem saves recurrent state where consecutive prompts diverge, allowing later requests to reuse previously processed context. It keeps most prefill on the matrix unit and produces rollback states through a short per-token tail. Its verification kernel reads each weight block once for all proposed tokens. Between request bursts, Tandem releases NPU power votes without unloading the model. We implement Tandem in llama.cpp and evaluate it on recorded agent interactions across two Snapdragon generations. On a Snapdragon 8 Elite, Tandem reduces per-step latency by 30% for text-based screen descriptions and 16% for screenshots relative to llama.cpp's fastest NPU configuration, reduces energy per text step by 33%, and lowers standby power with the model loaded from 0.76 to 0.22 W.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Latency (audio), Operating system, Decoding methods, Reuse, Rollback

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange