# GUIDE: Interpretable GUI Agent Evaluation via Hierarchical Diagnosis

arXiv: https://arxiv.org/abs/2604.04399 · HTML: https://mlsanity.com/paper/2604.04399/

## Key facts

- Cited by 7 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 15 of its 25 references that are in the map.
- First posted on arXiv 6 Apr 2026.

## Placed in

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

## Builds on (most cited first, 15 of 15)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [Mind2Web: Towards a Generalist Agent for the Web](https://mlsanity.com/paper/2306.06070/) (2023)
- [GPT-4V(ision) is a Generalist Web Agent, if Grounded](https://mlsanity.com/paper/2401.01614/) (2024)
- [AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents](https://mlsanity.com/paper/2401.13178/) (2024)
- [An Illusion of Progress? Assessing the Current State of Web Agents](https://mlsanity.com/paper/2504.01382/) (2025)
- [Autonomous Evaluation and Refinement of Digital Agents](https://mlsanity.com/paper/2404.06474/) (2024)
- [AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories](https://mlsanity.com/paper/2504.08942/) (2025)
- [AgentPRM: Process Reward Models for LLM Agents via Step-Wise Promise and Progress](https://mlsanity.com/paper/2511.08325/) (2025)
- [Look Before You Leap: A GUI-Critic-R1 Model for Pre-Operative Error Diagnosis in GUI Automation](https://mlsanity.com/paper/2506.04614/) (2025)
- [BacktrackAgent: Enhancing GUI Agent with Error Detection and Backtracking Mechanism](https://mlsanity.com/paper/2505.20660/) (2025)
- [Auto-Eval Judge: Towards a General Agentic Framework for Task Completion Evaluation](https://mlsanity.com/paper/2508.05508/) (2025)
- [WebGraphEval: Multi-Turn Trajectory Evaluation for Web Agents using Graph Representation](https://mlsanity.com/paper/2510.19205/) (2025)

## Built on by (most cited first, 7 of 7)

- [CUADebug: Diagnosing and Repairing Computer-Use Agent Failures](https://mlsanity.com/paper/2608.02643/) (2026)
- [Are Android GUI Agents Robust Against Runtime Anomalies? AnTrap: Evaluating Agents in Dynamic Adversarial Environments](https://mlsanity.com/paper/2608.24099/) (2026)
- [Securing Computer-Use Agents: A Unified Architecture-Lifecycle Framework for Deployment-Grounded Reliability](https://mlsanity.com/paper/2605.07110/) (2026)
- [DENSE: Distilling Agent Trajectories into Evidence-Grounded Shortcut Trees for Self-Refinement](https://mlsanity.com/paper/2609.21423/) (2026)
- [GUITAR: Structured Failure Diagnosis of GUI Agents via State Transitions](https://mlsanity.com/paper/2609.34113/) (2026)
- [Same Tasks, Different Apps: Why Mobile GUI Agents Fail to Generalize?](https://mlsanity.com/paper/2609.34139/) (2026)
- [Adaptive-Group Process Rewards for GUI Agents](https://mlsanity.com/paper/2610.17710/) (2026)

## Abstract (arXiv)

Evaluating GUI agents presents a distinct challenge: trajectories are long, visually grounded, and open-ended, yet evaluation must be both accurate and interpretable. Existing approaches typically apply a single holistic judgment over the entire action-observation sequence-a strategy that proves unreliable on long-horizon tasks and yields binary verdicts offering no insight into where or why an agent fails. This opacity limits the utility of evaluation as a diagnostic tool for agent development. We introduce GUIDE (GUI Understanding and Interpretable Diagnostic Evaluation), a framework that decomposes trajectory assessment into three sequential stages mirroring the compositional structure of GUI tasks. Trajectory Segmentation partitions the full trace into semantically coherent subtask units. Subtask Diagnosis evaluates each unit in context, assigning a completion verdict and generating a structured error analysis with corrective recommendations. Overall Summary aggregates per-subtask diagnoses into a task-level judgment. By operating on bounded subtask segments rather than full trajectories, GUIDE mitigates the context overload that degrades existing evaluators as task complexity grows. We validate GUIDE on three benchmarks: an industrial e-commerce dataset of 932 trajectories, AGENTREWARDBENCH spanning five web agent tasks with 1302 trajectories, and AndroidBench for mobile device control. Across all settings, GUIDE substantially outperforms existing evaluators-achieving up to 5.35 percentage points higher accuracy than the strongest baseline-while producing structured diagnostic reports that directly inform agent improvement.