# Understanding the Weakness of Large Language Model Agents within a Complex Android Environment

arXiv: https://arxiv.org/abs/2402.06596 · HTML: https://mlsanity.com/paper/2402.06596/

## Key facts

- Cited by 60 papers in the map (data through 14 Oct 2026).
- Placed in GUI and mobile agents, one of 16 directions in LLM agents and tool use.
- Builds on 21 of its 42 references that are in the map.
- First posted on arXiv 9 Feb 2024.

## Placed in

- LLM agents and tool use / [GUI and mobile agents](https://mlsanity.com/agents/gui-and-mobile-agents/)

## Builds on (most cited first, 21 of 21)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [A Survey on Large Language Model based Autonomous Agents](https://mlsanity.com/paper/2308.11432/) (2023)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [The Rise and Potential of Large Language Model Based Agents: A Survey](https://mlsanity.com/paper/2309.07864/) (2023)
- [Mind2Web: Towards a Generalist Agent for the Web](https://mlsanity.com/paper/2306.06070/) (2023)
- [ALFWorld: Aligning Text and Embodied Environments for Interactive Learning](https://mlsanity.com/paper/2010.03768/) (2020)
- [WizardLM: Empowering large pre-trained language models to follow complex instructions](https://mlsanity.com/paper/2304.12244/) (2023)
- [Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs](https://mlsanity.com/paper/2305.03111/) (2023)
- [Android in the Wild: A Large-Scale Dataset for Android Device Control](https://mlsanity.com/paper/2307.10088/) (2023)
- [Contextual Markov Decision Processes](https://mlsanity.com/paper/1502.02259/) (2015)
- [GPT-4V in Wonderland: Large Multimodal Models for Zero-Shot Smartphone GUI Navigation](https://mlsanity.com/paper/2311.07562/) (2023)
- [SmartPlay: A Benchmark for LLMs as Intelligent Agents](https://mlsanity.com/paper/2310.01557/) (2023)
- [Can ChatGPT Defend its Belief in Truth? Evaluating LLM Reasoning via Debate](https://mlsanity.com/paper/2305.13160/) (2023)
- [AndroidEnv: A Reinforcement Learning Platform for Android](https://mlsanity.com/paper/2105.13231/) (2021)
- [LASER: LLM Agent with State-Space Exploration for Web Navigation](https://mlsanity.com/paper/2309.08172/) (2023)
- [Agent Instructs Large Language Models to be General Zero-Shot Reasoners](https://mlsanity.com/paper/2310.03710/) (2023)

## Built on by (most cited first, 30 of 60)

- [AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agents](https://mlsanity.com/paper/2405.14573/) (2024)
- [AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents](https://mlsanity.com/paper/2503.18666/) (2025)
- [GUIOdyssey: A Comprehensive Dataset for Cross-App GUI Navigation on Mobile Devices](https://mlsanity.com/paper/2406.08451/) (2024)
- [BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games](https://mlsanity.com/paper/2411.13543/) (2024)
- [AndroidLab: Training and Systematic Benchmarking of Android Autonomous Agents](https://mlsanity.com/paper/2410.24024/) (2024)
- [SPA-Bench: A Comprehensive Benchmark for SmartPhone Agent Evaluation](https://mlsanity.com/paper/2410.15164/) (2024)
- [MobileAgentBench: An Efficient and User-Friendly Benchmark for Mobile LLM Agents](https://mlsanity.com/paper/2406.08184/) (2024)
- [A Survey on (M)LLM-Based GUI Agents](https://mlsanity.com/paper/2504.13865/) (2025)
- [A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agents](https://mlsanity.com/paper/2601.12538/) (2026)
- [Code as Agent Harness](https://mlsanity.com/paper/2605.18747/) (2026)
- [Magentic-UI: Towards Human-in-the-loop Agentic Systems](https://mlsanity.com/paper/2507.22358/) (2025)
- [CRAB: Cross-environment Agent Benchmark for Multimodal Language Model Agents](https://mlsanity.com/paper/2407.01511/) (2024)
- [Advancing Mobile GUI Agents: A Verifier-Driven Approach to Practical Deployment](https://mlsanity.com/paper/2503.15937/) (2025)
- [MCPWorld: A Unified Benchmarking Testbed for API, GUI, and Hybrid Computer Use Agents](https://mlsanity.com/paper/2506.07672/) (2025)
- [LlamaTouch: A Faithful and Scalable Testbed for Mobile UI Task Automation](https://mlsanity.com/paper/2404.16054/) (2024)
- [OS Agents: A Survey on MLLM-based Agents for General Computing Devices Use](https://mlsanity.com/paper/2508.04482/) (2025)
- [AutoDroid-V2: Boosting SLM-based GUI Agents via Code Generation](https://mlsanity.com/paper/2412.18116/) (2024)
- [LLM-Powered GUI Agents in Phone Automation: Surveying Progress and Prospects](https://mlsanity.com/paper/2504.19838/) (2025)
- [From Language to Action: A Review of Large Language Models as Autonomous Agents and Tool Users](https://mlsanity.com/paper/2508.17281/) (2025)
- [Foundations and Recent Trends in Multimodal Mobile Agents: A Survey](https://mlsanity.com/paper/2411.02006/) (2024)
- [VeriSafe Agent: Safeguarding Mobile GUI Agent via Logic-based Action Verification](https://mlsanity.com/paper/2503.18492/) (2025)
- [FingerTip 20K: A Benchmark for Proactive and Personalized Mobile LLM Agents](https://mlsanity.com/paper/2507.21071/) (2025)
- [A3: Android Agent Arena for Mobile GUI Agents with Essential-State Procedural Evaluation](https://mlsanity.com/paper/2501.01149/) (2025)
- [MobA: Multifaceted Memory-Enhanced Adaptive Planning for Efficient Mobile Task Automation](https://mlsanity.com/paper/2410.13757/) (2024)
- [The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestration](https://mlsanity.com/paper/2603.22862/) (2026)
- [Evolutionary Perspectives on the Evaluation of LLM-Based AI Agents: A Comprehensive Survey](https://mlsanity.com/paper/2506.11102/) (2025)
- [AI Gamestore: Scalable, Open-Ended Evaluation of Machine General Intelligence with Human Games](https://mlsanity.com/paper/2602.17594/) (2026)
- [MAS-Bench: A Unified Benchmark for Shortcut-Augmented Hybrid Mobile GUI Agents](https://mlsanity.com/paper/2509.06477/) (2025)
- [AndroidGen: Building an Android Language Agent under Data Scarcity](https://mlsanity.com/paper/2504.19298/) (2025)
- [Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents](https://mlsanity.com/paper/2508.19504/) (2025)

## Abstract (arXiv)

Large language models (LLMs) have empowered intelligent agents to execute intricate tasks within domain-specific software such as browsers and games. However, when applied to general-purpose software systems like operating systems, LLM agents face three primary challenges. Firstly, the action space is vast and dynamic, posing difficulties for LLM agents to maintain an up-to-date understanding and deliver accurate responses. Secondly, real-world tasks often require inter-application cooperation}, demanding farsighted planning from LLM agents. Thirdly, agents need to identify optimal solutions aligning with user constraints, such as security concerns and preferences. These challenges motivate AndroidArena, an environment and benchmark designed to evaluate LLM agents on a modern operating system. To address high-cost of manpower, we design a scalable and semi-automated method to construct the benchmark. In the task evaluation, AndroidArena incorporates accurate and adaptive metrics to address the issue of non-unique solutions. Our findings reveal that even state-of-the-art LLM agents struggle in cross-APP scenarios and adhering to specific constraints. Additionally, we identify a lack of four key capabilities, i.e., understanding, reasoning, exploration, and reflection, as primary reasons for the failure of LLM agents. Furthermore, we provide empirical analysis on the failure of reflection, and improve the success rate by 27% with our proposed exploration strategy. This work is the first to present valuable insights in understanding fine-grained weakness of LLM agents, and offers a path forward for future research in this area. Environment, benchmark, and evaluation code for AndroidArena are released at https://github.com/AndroidArenaAgent/AndroidArena.