"AIMH 2.0 AI-based Management of Media, Textual, and Humanities Data 2.0" - Bando CR 010 2026 ISTI PI

id="offer-description" class="ecl-u-type-heading-2 ecl-u-mt-none ecl-u-mb-l">Offer Description The research activity will address artificial intelligence and deep learning applied to visual analysis and pattern recognition, focusing on two main directions. The first is class-agnostic counting, leveraging multimodal AI to develop models capable of counting objects of arbitrary categories, unseen during training, from a few visual exemplars or textual descriptions. This aims to create flexible and generalizable counting systems for open-world scenarios. The second explores bio-inspired learning approaches, such as Hebbian learning, as an alternative to backpropagation, and the use of biological neuronal cultures as a computational substrate. Across both directions, the activity will tackle the scarcity of annotated data through synthetic datasets generated in virtual environments, domain adaptation techniques, unsupervised learning, and strategies designed to handle noisy labels. Where to apply Website https://selezionionline.cnr.it/ . Requirements. Additional Information . Eligibility criteria. Eligible destination country/ies for fellows: . Italy. Eligibility of fellows: country/ies of residence: . Italy. EUROPE. OTHER. E.U. Eligibility of fellows: nationality/ies: . Italy. EUROPE. OTHER. E.U. Website for additional job details https://www.urp.cnr.it/node/54671. <h2