Lightweight Neural App Control
Filippos Christianos , Georgios Papoudakis , Thomas Coste , Jianye Hao , Jun Wang , Kun Shao
- 🏛 Institutions
- Huawei Noah's Ark Lab , UCL
- 📅 Date
- October 23, 2024
- 📑 Publisher
- ICLR 2025 (Spotlight)
- 💻 Env
- Mobile
- 🔑 Keywords
TLDR
Introduces LiMAC, a lightweight neural framework for Android app control that combines an Action Transformer with fine-tuned vision-language models. The paper reports large gains over prompt-only baselines on mobile control benchmarks while keeping the control stack compact.
Related papers (24)
- ClickAgent: Enhancing UI Location Capabilities of Autonomous AgentsOctober 9, 2024 · SIGDIAL 2025
- STaR-KV: Spatio-Temporal Adaptive Re-weighting for KV Cache Compression in GUI Vision-Language ModelsJune 1, 2026 · arXiv
- BlueLM-GUI Technical Report: A Real-Device-Centric Flywheel for Self-Improving Mobile GUI AgentsSeptember 11, 2026 · arXiv
- JarvisGUI: Towards Cross-Device GUI Agents with Dynamic Task CompositionSeptember 9, 2026 · arXiv
- APPSim-Bench: Bridging Real-world Apps and Reproducible Evaluation for Mobile GUI AgentsSeptember 7, 2026 · arXiv
- Improving Proficiency and Efficiency of Android GUI Agents via Self-Generating Tool ActionsSeptember 6, 2026 · arXiv
- ElderBench: Benchmarking Autonomous Mobile Agents for Older AdultsSeptember 4, 2026 · arXiv
- WiP: Characterizing and Defending Against Mobile-Agent-Driven MFA AutomationSeptember 2, 2026 · arXiv
- GUI-CC: Benchmarking Contextual Consistency of GUI World Models as Agent EnvironmentsAugust 30, 2026 · arXiv
- ActReal: System-Level Mobile Agents Challenge Mobile Automation DetectionAugust 30, 2026 · arXiv
- WM-R1: Training GUI Agents to Reason and leverage World Models with Reinforcement LearningAugust 27, 2026 · arXiv
- Are Android GUI Agents Robust Against Runtime Anomalies? AnTrap: Evaluating Agents in Dynamic Adversarial EnvironmentsAugust 25, 2026 · arXiv
- ADeptS-Bench: Measuring the Trustworthiness of Computer Use Agents Across DevicesAugust 25, 2026 · arXiv
- GSAR: Goal-State-Anchor Rewards for Mobile GUI Agents with Self-Evolving Data SynthesisAugust 24, 2026 · arXiv
- Lexical Coupling in GUI Element Grounding: Sentence Embeddings Track Labels across Mobile and WebAugust 22, 2026 · arXiv
- Benchmarking General Mobile Assistants in Challenging Real-World ScenariosAugust 21, 2026 · arXiv
- Automated Trajectory Evaluation for Mobile Agents via Step-Level Consequence Reasoning and AggregationAugust 21, 2026 · arXiv
- MobileWorldSafety: Benchmarking GUI Agent Safety Against Environmental Injection Attacks in Android AppsAugust 18, 2026 · arXiv
- CoAdapt-GUI: Joint Workflow Context and Policy Adaptation for Unseen GUI ApplicationsAugust 12, 2026 · arXiv
- The Next Screenshot Knows: Gated Hindsight Distillation for Mobile GUI AgentsAugust 6, 2026 · arXiv
- StepReflect: Structured UI Transition Reflection for Mobile GUI AgentsAugust 6, 2026 · arXiv
- AppDeltaWorld: Transition-Grounded Delta Code World Model for Mobile GUI AgentsAugust 6, 2026 · arXiv
- MAGA: Multi-Platform Self-Fusion of GUI Agents via Structured Action DistillationJuly 31, 2026 · arXiv
- Qwen-UI-Agent Technical Report: Toward Next-Generation Real-World Centric Foundation GUI AgentsJuly 30, 2026 · arXiv