Applied Researcher

David La Barbera

Research rigor. Engineering execution.

I develop research-level AI systems where theory meets reality.

RAG AI agents Human-in-the-loop IR Data Science

Projects Spotlight

LLMPatient

Private build

Can we let an AI-agent simulate a mentally ill patient?

Dual-memory (short/long-term) · multi-session continuity · text + voice.

LLM agentsMemorySafetyEvaluation
Project Code: private

FVG Meteo Analysis

How does the climate changes in FVG region?

Month-to-month similarity · anomaly detection · dashboard in progress.

Time series Anomaly detection Python Visualization
GitHub

Human–AI Collaboration in Fact-Checking

How to let people collaborate with AI in performing fact-checking?

Bias-aware HITL pipelines · optimized crowdsourcing · human-AI collaboration.

Human-in-the-loop Evaluation RAG human/AI bias
PhD Thesis (PDF)
See more →

Timeline

  1. 2025-Now
    Postdoctoral Researcher · University of Milano-Bicocca
    LLM-driven therapeutic simulation · evaluation & failure-mode analysis
    AgentsSafetyEvaluation
  2. 2025-Now
    Contract Professor · University of Pavia
    Information Retrieval & Recommender Systems (BSc AI)
    IRTeaching
  3. 2021-2025
    PhD · Excellent cum Laude
    Computer Science & AI · University of Udine, Italy
    HITLRAGBiasEvaluation
  4. 2016-2019
    MSc · 110/110 cum Laude
    Computer Science · University of Udine, Italy
    NLPIRData Science
  5. 2012-2016
    BSc · 96/110
    Multimedia and Web Technologies · University of Udine, Italy
    CSWeb DevEngineering

Publications

Peer-reviewed papers and selected technical reports (sorted by year).

A Comparative Analysis of Retrieval-Augmented Generation and Crowdsourcing for Fact-Checking ECIR · 2025
Impersonating the Crowd: Evaluating LLMs' Ability to Replicate Human Judgment in Misinformation Assessment ICTIR · 2025 Best Paper
The Magnitude of Truth: On Using Magnitude Estimation for Truthfulness Assessment SIGIR · 2025
Cognitive Biases in Fact-Checking and Their Countermeasures: A Review Information Processing & Management · 2024
Crowdsourced Fact-checking: Does It Actually Work? Information Processing & Management · 2024
Supporting Fair and Efficient Emergency Medical Services in a Large Heterogeneous Region Journal of Healthcare Informatics Research · 2024
Report on the 14th Italian Information Retrieval Workshop (IIR 2024) SIGIR Forum · 2024
Report on the Hands-On PhD Course on Responsible AI from the Lens of an Information Access Researcher SIGIR Forum · 2024
How Many Crowd Workers Do I Need? On Statistical Power when Crowdsourcing Relevance Judgments ACM Transactions on Information Systems · 2024
The Elusiveness of Detecting Political Bias in Language Models CIKM · 2024
Enhancing Fact-Checking: From Crowdsourced Validation to Integration with Large Language Models IIR · 2024
Combining Large Language Models and Crowdsourcing for Hybrid Human-AI Misinformation Detection SIGIR · 2024
Combining human intelligence and machine learning for fact-checking: Towards a hybrid human-in-the-loop framework Intelligenza Artificiale · 2023
Combining Human and Machine Confidence in Truthfulness Assessment ACM Journal of Data and Information Quality · 2023
Fact-Checking at Scale with Crowdsourcing: Experiments and Lessons Learned IIR · 2023
HEROHE Challenge: Predicting HER2 Status in Breast Cancer from Hematoxylin-Eosin Whole-Slide Imaging Journal of Imaging · 2022
A Hybrid Human-In-The-Loop Framework for Fact Checking NL4AI@AI*IA · 2022
A Multi-objective Biased Random-Key Genetic Algorithm for the Siting of Emergency Vehicles MIC · 2022
BUM at CheckThat!-2022: A Composite Deep Learning Approach to Fake News Detection using Evidence Retrieval CLEF (Working Notes) · 2022
The Effects of Crowd Worker Biases in Fact-Checking Tasks FAccT · 2022
The many dimensions of truthfulness: Crowdsourcing misinformation assessments on a multidimensional scale Information Processing & Management · 2021
A Software Simulator for Optimizing Ambulance Location and Response Time: A Preliminary Report IEEE ICDH · 2021
The Many Dimensions of Truthfulness: Crowdsourcing Misinformation Assessments on a Multidimensional Scale arXiv · 2021
HEROHE Challenge: assessing HER2 status in breast cancer without immunohistochemistry or in situ hybridization arXiv · 2021
Detection of HER2 from Haematoxylin-Eosin Slides Through a Cascade of Deep Learning Classifiers via Multi-Instance Learning Journal of Imaging · 2020
Crowdsourcing Truthfulness: The Impact of Judgment Scale and Assessor Bias ECIR · 2020

Contact

Email LinkedIn