# Visual Test-time Scaling for GUI Agent Grounding

arXiv: https://arxiv.org/abs/2505.00684 · HTML: https://mlsanity.com/paper/2505.00684/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 21 of its 47 references that are in the map.
- First posted on arXiv 1 May 2025.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)
- 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)

- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [WebGPT: Browser-assisted question-answering with human feedback](https://mlsanity.com/paper/2112.09332/) (2021)
- [UI-TARS: Pioneering Automated GUI Interaction with Native Agents](https://mlsanity.com/paper/2501.12326/) (2025)
- [SeeClick: Harnessing GUI Grounding for Advanced Visual GUI Agents](https://mlsanity.com/paper/2401.10935/) (2024)
- [GPT-4V(ision) is a Generalist Web Agent, if Grounded](https://mlsanity.com/paper/2401.01614/) (2024)
- [WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models](https://mlsanity.com/paper/2401.13919/) (2024)
- [Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents](https://mlsanity.com/paper/2410.05243/) (2024)
- [Mobile-Agent: Autonomous Multi-Modal Mobile Device Agent with Visual Perception](https://mlsanity.com/paper/2401.16158/) (2024)
- [ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use](https://mlsanity.com/paper/2504.07981/) (2025)
- [PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs](https://mlsanity.com/paper/2402.07872/) (2024)
- [You Only Look at Screens: Multimodal Chain-of-Action Agents](https://mlsanity.com/paper/2309.11436/) (2023)
- [GPT-4V in Wonderland: Large Multimodal Models for Zero-Shot Smartphone GUI Navigation](https://mlsanity.com/paper/2311.07562/) (2023)
- [VisualAgentBench: Towards Large Multimodal Models as Visual Foundation Agents](https://mlsanity.com/paper/2408.06327/) (2024)
- [ExACT: Teaching AI Agents to Explore with Reflective-MCTS and Exploratory Learning](https://mlsanity.com/paper/2410.02052/) (2024)
- [All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuning](https://mlsanity.com/paper/2503.01067/) (2025)
- [VisualWebArena: Evaluating Multimodal Agents on Realistic Visual Web Tasks](https://mlsanity.com/paper/2401.13649/) (2024)
- [OmniParser V2: Structured-Points-of-Thought for Unified Visual Text Parsing and Its Generality to Multimodal Large Language Models](https://mlsanity.com/paper/2502.16161/) (2025)
- [Vision-Language Model for Object Detection and Segmentation: A Review and Evaluation](https://mlsanity.com/paper/2504.09480/) (2025)
- [Learning what and where to attend](https://mlsanity.com/paper/1805.08819/) (2018)

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

- [DiMo-GUI: Advancing Test-time Scaling in GUI Grounding via Modality-Aware Visual Reasoning](https://mlsanity.com/paper/2507.00008/) (2025)
- [Phi-Ground Tech Report: Advancing Perception in GUI Grounding](https://mlsanity.com/paper/2507.23779/) (2025)
- [Test-Time Reinforcement Learning for GUI Grounding via Region Consistency](https://mlsanity.com/paper/2508.05615/) (2025)
- [SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agents](https://mlsanity.com/paper/2506.21669/) (2025)
- [Hybrid Self-evolving Structured Memory for GUI Agents](https://mlsanity.com/paper/2603.10291/) (2026)
- [MVP: Multiple View Prediction Improves GUI Grounding](https://mlsanity.com/paper/2512.08529/) (2025)
- [Zoom in, Click out: Unlocking and Evaluating the Potential of Zooming for GUI Grounding](https://mlsanity.com/paper/2512.05941/) (2025)
- [UI-Zoomer: Uncertainty-Driven Adaptive Zoom-In for GUI Grounding](https://mlsanity.com/paper/2604.14113/) (2026)
- [AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents](https://mlsanity.com/paper/2605.17933/) (2026)
- [Zoom to Essence: Trainless GUI Grounding by Inferring upon Interface Elements](https://mlsanity.com/paper/2603.14448/) (2026)
- [Measure Twice, Click Once: Co-evolving Proposer and Visual Critic via Reinforcement Learning for GUI Grounding](https://mlsanity.com/paper/2604.21268/) (2026)
- [Generalist Scanner Meets Specialist Locator: A Synergistic Coarse-to-Fine Framework for Robust GUI Grounding](https://mlsanity.com/paper/2509.24133/) (2025)
- [Zoom Consistency: A Free Confidence Signal in Multi-Step Visual Grounding Pipelines](https://mlsanity.com/paper/2604.15376/) (2026)
- [What Happens Before Decoding? Prefill Determines GUI Grounding in VLMs](https://mlsanity.com/paper/2605.12549/) (2026)
- [MEGA-GUI: Multi-stage Enhanced Grounding Agents for GUI Elements](https://mlsanity.com/paper/2511.13087/) (2025)
- [GUI Agents with Reinforcement Learning: Toward Digital Inhabitants](https://mlsanity.com/paper/2604.27955/) (2026)
- [AutoFocus: Uncertainty-Aware Active Visual Search for GUI Grounding](https://mlsanity.com/paper/2605.02630/) (2026)
- [WebRouter: Query-specific Router via Variational Information Bottleneck for Cost-sensitive Web Agent](https://mlsanity.com/paper/2510.11221/) (2025)
- [One Forward Beats Two: InnerZoom for Accurate and Efficient GUI Grounding](https://mlsanity.com/paper/2606.30084/) (2026)
- [Test-Time Self-Evolving GUI Visual Grounding via Reflection-Guided On-Policy Self-Distillation](https://mlsanity.com/paper/2608.11191/) (2026)
- [WebChallenger: A Reliable and Efficient Generalist Web Agent](https://mlsanity.com/paper/2606.10423/) (2026)
- [Thinking with Patterns: Breaking the Perceptual Bottleneck in Visual Planning via Pattern Induction](https://mlsanity.com/paper/2605.16848/) (2026)
- [LookAgain: Closed-Loop GUI Grounding with Visually Grounded Reflection](https://mlsanity.com/paper/2608.09723/) (2026)
- [GUI-Primitives: Diagnosing Spatial Reasoning Failures in Vision-Language GUI Grounding](https://mlsanity.com/paper/2608.21832/) (2026)
- [InSight: A Benchmark for Agentic Claim Verification in Interactive Visualizations](https://mlsanity.com/paper/2609.01383/) (2026)
- [Efficient GUI Agents: A Systems Survey of Observation, Memory, Action, and Runtime Optimization](https://mlsanity.com/paper/2609.02309/) (2026)
- [ScreenHaystack: Finding Blind Zones in GUI Grounding](https://mlsanity.com/paper/2609.32036/) (2026)
- [Recurrent Latent Visual Search for GUI Grounding](https://mlsanity.com/paper/2610.05185/) (2026)
- [ConClick: Element-Wise Contrastive Learning Improves GUI Grounding](https://mlsanity.com/paper/2610.13268/) (2026)
- [MHVF: Training-Free GUI Grounding via Verification-Guided Multi-Hypothesis Search](https://mlsanity.com/paper/2610.17792/) (2026)

## Abstract (arXiv)

We introduce RegionFocus, a visual test-time scaling approach for Vision Language Model Agents. Understanding webpages is challenging due to the visual complexity of GUI images and the large number of interface elements, making accurate action selection difficult. Our approach dynamically zooms in on relevant regions, reducing background clutter and improving grounding accuracy. To support this process, we propose an image-as-map mechanism that visualizes key landmarks at each step, providing a transparent action record and enables the agent to effectively choose among action candidates. Even with a simple region selection strategy, we observe significant performance gains of 28+% on Screenspot-pro and 24+% on WebVoyager benchmarks on top of two state-of-the-art open vision language model agents, UI-TARS and Qwen2.5-VL, highlighting the effectiveness of visual test-time scaling in interactive settings. We achieve a new state-of-the-art grounding performance of 61.6% on the ScreenSpot-Pro benchmark by applying RegionFocus to a Qwen2.5-VL-72B model. Our code will be released publicly at https://github.com/tiangeluo/RegionFocus.