Sam Sterling

Vision-Language Models for Image-Based Dietary Assessment: A Benchmark of Accuracy, Cost, and Prompt Strategies Across Ten Models

This study benchmarks ten vision-language models for automated food recognition, weight, and calorie estimation using the Nutrition5k dataset, finding Gemini 3.0 Flash and 3.1 Flash Lite as top performers.

A multi-scale segmentation-free self-supervised AI model to characterize the heterogeneity of the brain tumor microenvironment

This paper introduces a segmentation-free self-supervised AI model to analyze brain tumor microenvironments, identifying a novel tumor-associated B cell subset linked to improved survival.