Information Systems Lab @ Polytechnic University of Bari

Unifying Academic Rigor and Industrial Scale for Responsible, Reproducible, and Efficient Recommendation

A comprehensive collection of works published in the latest literature about Information Retrieval and Recommender Systems focused on their increasingly symbiotic evolution through Large Language Models.

ABAI Hands-on Sessions Repository

STABLE-BAG: a Stability-Aware Framework for SHAP-based Brain Age Gap Decomposition

Code for the paper "PUMA: Post-Hoc Sparsification of Universal Multimodal Embeddings for Efficient Retrieval" accepted to EMNLP 2026 (Findings).

Comprehensive and Rigorous Framework for Reproducible Recommender Systems Evaluation

Now You Have My Healthy Attention: A U-DiT for Brain-MRI Inpainting - Accepted at MICCAI 2026 Challenge BraTS Inpainting

WaveDiT: Distribution-Aware Wavelet Flow Matching for Efficient 3D Brain MRI Synthesis - Accepted at MICCAI 2026

cellular direct search