vcalib: differentiable calibration filters for RF-DETR under illumination shift

Sat, 04 Jul 2026 00:00:00 +0000

When the lighting changes, a detector trained under other conditions degrades. The textbook answer is to collect data, label it, retrain and redeploy. vcalib explores the cheap answer: keep the model frozen and fit, in minutes and without labels, a tiny filter that corrects the input.

The problem

RF-DETR is a state-of-the-art real-time detector and, like almost any vision model, it is sensitive to illumination: dawn vs. dusk, indoor vs. overcast, a new camera, a new site. When the input distribution drifts away from the training one, detection quality drops. Retraining or fine-tuning is expensive, needs labels and is impractical on edge devices.

Misinformation detection in fiction works with LLM-generated ontologies

Sun, 15 Jun 2025 00:00:00 +0000

Most anti-misinformation tools target fake news. Films, series and podcasts blend facts with creative license, and the brain does not always tell them apart: people are 40% more likely to believe a false claim when it arrives visually, and once absorbed it keeps influencing them about half of the time even after a correction.

Context

Work carried out at the BISITE research group (University of Salamanca) within the TRUESTORIES project (CPP2021-008358), funded by MICIU/AEI and the European Union (NextGenerationEU/PRTR), and published at DCAI 2025, the International Conference on Distributed Computing and Artificial Intelligence: AI-Powered Ontology-Based Architecture for Misinformation Detection in Fiction Works.

Projects on

vcalib: differentiable calibration filters for RF-DETR under illumination shift

The problem

Misinformation detection in fiction works with LLM-generated ontologies

Context