# GUI-Primitives: Diagnosing Spatial Reasoning Failures in Vision-Language GUI Grounding

arXiv: https://arxiv.org/abs/2608.21832 · HTML: https://mlsanity.com/paper/2608.21832/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 17 of its 33 references that are in the map.
- First posted on arXiv 22 Aug 2026.

## Placed in

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

## Builds on (most cited first, 17 of 17)

- [Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs](https://mlsanity.com/paper/2406.16860/) (2024)
- [SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities](https://mlsanity.com/paper/2401.12168/) (2024)
- [UI-TARS: Pioneering Automated GUI Interaction with Native Agents](https://mlsanity.com/paper/2501.12326/) (2025)
- [BLINK: Multimodal Large Language Models Can See but Not Perceive](https://mlsanity.com/paper/2404.12390/) (2024)
- [Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents](https://mlsanity.com/paper/2410.05243/) (2024)
- [OS-ATLAS: A Foundation Action Model for Generalist GUI Agents](https://mlsanity.com/paper/2410.23218/) (2024)
- [Set-of-Mark Prompting Unleashes Extraordinary Visual Grounding in GPT-4V](https://mlsanity.com/paper/2310.11441/) (2023)
- [Visual Spatial Reasoning](https://mlsanity.com/paper/2205.00363/) (2022)
- [What's "up" with vision-language models? Investigating their struggle with spatial reasoning](https://mlsanity.com/paper/2310.19785/) (2023)
- [GTA1: GUI Test-time Scaling Agent](https://mlsanity.com/paper/2507.05791/) (2025)
- [Your Large Vision-Language Model Only Needs A Few Attention Heads For Visual Grounding](https://mlsanity.com/paper/2503.06287/) (2025)
- [MMBench-GUI: Hierarchical Multi-Platform Evaluation Framework for GUI Agents](https://mlsanity.com/paper/2507.19478/) (2025)
- [Beyond Semantics: Rediscovering Spatial Awareness in Vision-Language Models](https://mlsanity.com/paper/2503.17349/) (2025)
- [List Items One by One: A New Data Source and Learning Paradigm for Multimodal LLMs](https://mlsanity.com/paper/2404.16375/) (2024)
- [Visual Test-time Scaling for GUI Agent Grounding](https://mlsanity.com/paper/2505.00684/) (2025)
- [Learning GUI Grounding with Spatial Reasoning from Visual Feedback](https://mlsanity.com/paper/2509.21552/) (2025)
- [Understanding Space Is Rocket Science -- Only Top Reasoning Models Can Solve Spatial Understanding Tasks](https://mlsanity.com/paper/2509.02175/) (2025)

## Abstract (arXiv)

Computer-use agents ground natural-language instructions in screenshots to locate interface elements, yet existing benchmarks do not isolate whether models bind relational language to the correct element. We introduce GUI-Primitives, a 994-item benchmark of contrastive instruction pairs over seven spatial relations in graphical user interfaces (left/right, above/below, containment, alignment, proximity, list ordinal, occlusion). Each pair holds the screenshot and anchor fixed while changing the relation expression, so the correct target moves between two designated candidates. Five annotators validate a 196-item subset (kappa = 0.94 well-formedness; kappa = 0.79 target selection). Nineteen vision-language models reach at most 32% strict point-in-box accuracy. Because models emit unconstrained coordinates, we classify each prediction by the candidate region it falls within. Predictions fall outside both candidates on 60-92% of items. Conditional on falling within a candidate region, target selection reaches 0.82-0.90 for horizontal position, vertical position, proximity, and list ordinal, but does not differ significantly from 0.50 for containment and occlusion: most failures reflect candidate localization rather than relation understanding. Across ten models, benchmark accuracy correlates with ScreenSpot-Pro accuracy (Spearman ρ = +0.74), an exploratory association at this sample size. Marking the two designated candidates raises selection accuracy by 35--57 percentage points, an oracle diagnostic that supplies the candidate set rather than a deployable method. We release the benchmark, predictions, and code.