YOLO with Kolmogorov-Arnold networks and vision-language foundation models for interpretable object detection with trustworthy multimodal AI in computer vision perception
Marios Impraimakis, Daniel Vazquez, Feiyu Zhou · 2026
Read YOLO with Kolmogorov-Arnold networks and vision-language foundation models for interpretable object detection with trustworthy multimodal AI in computer vision perception by Marios Impraimakis, Daniel Vazquez, Feiyu Zhou (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13538510.txt.
Description
Marios Impraimakis, Daniel Vazquez, and Feiyu Zhou add a trust layer to fast object detection in a 2026 study. Detectors label objects quickly yet hide whether each confidence score deserves trust in blurry or blocked scenes. A curve-based network modeled trust from seven shape and meaning features tied to each detection, keeping every feature influence visible. A vision-language model also captioned scenes without disturbing the trust layer.
Tests on a large public photo set and campus images showed the layer correctly flagging low-trust calls under blur, blocking, or weak texture. Smooth feature maps revealed exactly when confidence stood on solid ground and when it wobbled. The authors offer the system as a transparent perception piece for self-driving and mixed-media tools that must justify acceptance or review.
Catalog record
| CREATOR | Marios Impraimakis, Daniel Vazquez, Feiyu Zhou |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39536 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |