events

Nov 9 - Agentic Labeling for Specialized Visual Data

Uno studio su quando il prompting few-shot con modelli multimodali supera l’addestramento di un modello specializzato su dataset visivi specialistici, con un’illustrazione passo passo del workflow.

In inglese

Starts

2026-11-09T17:00:00+00:00

Organizer

Paris AI, Machine Learning and Computer Vision Meetup

Location

Online event

Run by the Paris AI, Machine Learning and Computer Vision Meetup, this session presents a study of multimodal models across 11 specialized visual datasets in manufacturing inspection, geology and medical imaging. It separates niche tasks from hard ones, reports how few labeled prompt examples shifted results on an industrial defect task, and notes where trained specialists pulled ahead. The talk covers building a few-shot labeling task, choosing reference examples and finding the crossover point between prompting and training. Participants get free GPU compute from Nebius to try the workflow.