How GUI Agents Fail: GraceBench for Execution Stability Diagnosis
cs.AI
Submitted: 2026-10-14
Updated: 2026-10-14
Code: https://github.com/ml-explore/mlx
Project page: https://iaie-physicalai.github.io/GraceBench
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Recovering Policy-Induced Errors: Benchmarking and Trajectory Synthesis for Robust GUI Agents
- A3: Android Agent Arena for Mobile GUI Agents with Essential-State Procedural Evaluation
- UI-KOBE: Knowledge-Oriented Behavior Exploration for Lightweight Graph-Guided GUI Agents
- SPA-Bench: A Comprehensive Benchmark for SmartPhone Agent Evaluation
- MobiFlow: Real-World Mobile Agent Benchmarking through Trajectory Fusion
- Agent-SAMA: State-Aware Mobile Assistant
- On the Effects of Data Scale on UI Control Agents
- ColorBench: Benchmarking Mobile Agents with Graph-Structured Framework for Complex Long-Horizon Tasks
- AndroidDaily: A Verifiable Benchmark for Mobile GUI Agents on Real-World Closed-Source Applications
- GUI-Xplore: Empowering Generalizable GUI Agents with One Exploration
- MobileVLM: A Vision-Language Model for Better Intra- and Inter-UI Understanding
- ProBench: Benchmarking GUI Agents with Accurate Process Information
- GraphPilot: GUI Task Automation with One-Step LLM Reasoning Powered by Knowledge Graph
- Large Language Model-Brained GUI Agents: A Survey
- ActionEngine: From Reactive to Programmatic Web Agents via State Machine Memory
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems