VRIQ: Benchmarking and Diagnosing the Visual-Reasoning IQ of Vision-Language Models
Vision-language model (VLM) errors on visual reasoning may come from missing visual structure, from mistakes after that structure is extracted, or from both. We introduce VRIQ, a benchmark of 1,391 human-verified IQ-style visual reasoning puzzles spanning five psychometric categories, each instantiated in category-matched abstract and natural-image domains. To study perception and reasoning more directly, we also introduce DiagVRIQ, a 209-item diagnostic subset paired with human-verified, rule-neutral textual descriptions. Across 21 VLMs, the 17 non-frontier models reach a 26.8% unweighted category mean on abstract puzzles, while three 2026 frontier models improve substantially but uneven...
| Authors | Tina Khezresmaeilzadeh, Jike Zhong, Konstantinos Psounis |
|---|---|
| Primary category | cs.CV |
| Published | 5 February 2026 |
| Last updated at source | 10 October 2026 |
Connected in the knowledge graph
- tina-khezresmaeilzadeh/VRIQDataset · cited by
Models and datasets referencing this paper
- tina-khezresmaeilzadeh/VRIQDataset
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.