Matched outcomes, divergent gaze: how foveated MLLMs search compared to humans

Mohamed Amine KERKOURI, Marouane Tliba, Aladine Chetouani, Ulas Bagci, Alessandro Bruno

ECCV 2026 (@ 3rd Workshop on Human-inspired Computer Vision)2026

Cite this work

Mohamed Amine KERKOURI, Marouane Tliba, Aladine Chetouani, Ulas Bagci, Alessandro Bruno (2026). Matched outcomes, divergent gaze: how foveated MLLMs search compared to humans. ECCV 2026 (@ 3rd Workshop on Human-inspired Computer Vision)

BibTeX

Cite this work

Related publications

Validating synthetically generated medical scans through human perceptual assessment & eye-tracking

GazeVaLM: a multi-observer eye-tracking benchmark for evaluating clinical realism in AI-generated x-rays

EyeAssist: how clinical context modulates expert gaze in volumetric reading and prognostic decision-making

NeoGaze: whose gaze counts? expertise, clinical information, and attention in neonatal radiology

GazeRefine: expert gaze as a test-time prompt for training-free medical image segmentation