Marco Postiglione

Marco Postiglione

Postdoctoral Research Scholar
Department of Computer Science, Northwestern University
email google scholar dblp linkedin
I develop trustworthy AI systems for social good, with applications spanning security and healthcare. My work focuses on bridging the gap between theoretical advances and deployed systems that serve real-world users, from fact-checkers and journalists to clinicians and public safety organizations. I am part of the Northwestern Security & AI Lab, led by Prof. V.S. Subrahmanian.

News

  • 2025 Paper "DEEP: A Discourse Evolution Engine for Predictions about Social Movements" has been accepted at the Thirty-Eighth Annual Conference on Innovative Applications of Artificial Intelligence (IAAI '26). Proceedings are available.
  • 2025 GODDS demo published at AAAI '25. Now serving 70+ news organizations worldwide.
  • 2025 Welcome! This site is now live — a small corner of the web to share my research, projects, and what I'm working on.

Selected Projects

GODDS
Global deepfake detection for 70+ newsrooms incl. PolitiFact & USA Today
SMART
Social movement analysis with WSJ, AP & Washington Post journalists
DEWS
Drone threat prediction deployed with Netherlands Police
NTEWS
Terrorism early warning system for national security
MNW Benchmark
Deepfake detection dataset with Microsoft AI for Good & WITNESS
Countering AI Proliferation
Executive training on countering AI proliferation threats

Selected Publications

* equal contribution · Full list on Google Scholar

Perturbed Public Voices (P²V): A Dataset for Robust Audio Deepfake Detection
C. Gao*, M. Postiglione*, I. Gortner, S. Kraus, V. S. Subrahmanian
Proceedings of the ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '26)
arxiv
MediCARE: Medical Collaborative Agents REasoning over Interpretable Heterogeneous Graphs
A. Ferraro*, A. Galli*, V. La Gatta*, M. Postiglione*, G. Riccio*, A. Romano*, G. M. Orlando*, D. Russo*, V. Moscato
Artificial Intelligence in Medicine, 2026
paper
ExpLusion: Explanation-driven Late Fusion for enhanced production process monitoring
A. Ferraro*, A. Galli*, M. Gallo*, V. La Gatta*, M. Postiglione*, V. Moscato
Computers & Industrial Engineering, vol. 217, 2026
paper
Brain3D: Brain Report Automation via Inflated Vision Transformers in 3D
M. Barone, F. Di Serio, G. Riccio, A. Romano, M. Postiglione, A. Ferraro, V. Moscato
Proceedings of the 39th IEEE International Symposium on Computer-Based Medical Systems (CBMS 2026) · Best Paper Award
arxiv
The Microsoft-Northwestern-WITNESS Benchmark for Deepfake Detection
T. Roca, M. Postiglione, C. Gao, I. Gortner, Z. Wojciak, P. Wang, M. Alimardani, S. Anlen, K. White, J. Lavista Ferres, S. Kraus, S. Gregory, V. S. Subrahmanian
IEEE Intelligent Systems, vol. 41, no. 2, 2026
paper dataset
SMART: A Social Movement Analysis & Reasoning Tool with Case Studies on #MeToo and #BlackLivesMatter
V. La Gatta*, M. Postiglione*, J. Gilbert, D. W. Linna Jr, M. M. Greenfield, A. Shaw, V. S. Subrahmanian
Proceedings of the ACM Web Conference 2026 (WWW '26), pp. 9634–9644
proceedings arxiv
AI-Generated Phishing: Combining Human Behavior with Post Content to Assess Susceptibility
N. Denisenko*, V. La Gatta*, M. Postiglione*, L. Sola*, Y. Chen, V. S. Subrahmanian
ACM Transactions on Internet Technology, 2026
paper
A Nonpartisan Study of Deepfake Activity and Engagement Around the 2024 US Presidential Election
M. Postiglione*, I. Gortner, L. Fosdick, C. Gao, S. Kraus, V. S. Subrahmanian
Proceedings of the International AAAI Conference on Web and Social Media (ICWSM '26)
DEEP: A Discourse Evolution Engine for Predictions about Social Movements
V. La Gatta*, M. Postiglione*, J. Gilbert, D. W. Linna Jr, M. M. Greenfield, A. Shaw, V. S. Subrahmanian
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI '26), IAAI Track
proceedings arxiv
SOLVE-Med: Specialized Orchestration for Leading Vertical Experts across Medical Specialties
R. Di Marino, G. Dioguardi, A. Romano, G. Riccio, M. Barone, M. Postiglione, F. Amato, V. Moscato
Proceedings of the 28th European Conference on Artificial Intelligence (ECAI 2025)
GODDS: The Global Online Deepfake Detection System
M. Postiglione*, J. Baldwin, N. Denisenko, L. Fosdick, C. Gao, I. Gortner, C. Pulice, S. Kraus, V. S. Subrahmanian
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI '25)
proceedings online system
Combining Evidence and Reasoning for Biomedical Fact-Checking
F. Barone*, A. Romano, A. Riccio, M. Postiglione, V. Moscato
Proceedings of the ACM SIGIR Conference on Research and Development in Information Retrieval, 2025
proceedings arxiv
Combating Biomedical Misinformation through Multi-modal Claim Detection and Evidence-based Verification
M. Barone, A. Romano, G. Riccio, M. Postiglione, V. Moscato
Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2025
A Drone Early Warning System for Predicting Threatening Trajectories
T. Deb, S. de Laaf, V. La Gatta, O. Lemmens, R. Lindelauf, M. van Meerten, H. Meerveld, A. Neeleman, M. Postiglione, V. S. Subrahmanian
IEEE Intelligent Systems, 2025
paper
PIE-Med: Predicting, Interpreting and Explaining Medical Recommendations
A. Romano, G. Riccio, M. Postiglione, V. Moscato
European Conference on Information Retrieval (ECIR 2025), Lecture Notes in Computer Science, vol. 15576, pp. 6–12
Predicting Future Disorders via Temporal Knowledge Graphs and Medical Ontologies
M. Postiglione*, D. Bean, Z. Kraljevic, R. Dobson, V. Moscato
IEEE Journal of Biomedical and Health Informatics, vol. 28, no. 7, 2024
paper
Explainability in AI-based Behavioral Malware Detection Systems
A. Galli*, V. La Gatta*, V. Moscato*, M. Postiglione*, G. Sperlì*
Computers & Security, vol. 141, 2024
paper
TaughtNet: Learning Multi-Task Biomedical NER from Single-Task Teachers
V. Moscato*, M. Postiglione*, C. Sansone*, G. Sperlì*
IEEE Journal of Biomedical and Health Informatics, vol. 27, no. 5, 2023
paper
Music Recommendation via Hypergraph Embedding
V. La Gatta*, V. Moscato*, M. Pennone*, M. Postiglione*, G. Sperlì*
IEEE Transactions on Neural Networks and Learning Systems, vol. 34, no. 10, 2022
paper
CASTLE: Cluster-Aided Space Transformation for Local Explanations
V. La Gatta*, V. Moscato*, M. Postiglione*, G. Sperlì*
Expert Systems with Applications, vol. 179, 2021
paper
An Epidemiological Neural Network Exploiting Dynamic Graph Structured Data Applied to COVID-19
V. La Gatta*, V. Moscato*, M. Postiglione*, G. Sperlì*
IEEE Transactions on Big Data, vol. 7, no. 1, 2020
paper

Experience & Education

Postdoctoral Research Scholar
Northwestern University · 2024–present
Northwestern Security & AI Lab (PI: V.S. Subrahmanian). Deepfake detection, disinformation, public safety AI.
Ph.D., Computer Science & Engineering
University of Naples Federico II · 2020–2024
Biomedical NLP, few-shot learning, temporal knowledge graphs. Advisor: Vincenzo Moscato.
Visiting Researcher
King's College London · 2023–2024
Dept. of Biostatistics & Health Informatics (PI: Richard Dobson).
M.S., Computer Science & Engineering
University of Naples Federico II · 2017–2020
Thesis: CASTLE — Cluster-Aided Space Transformation for Local Explanations. Advisor: Antonio Picariello.
B.S., Computer Science & Engineering
University of Naples Federico II · 2014–2017

Talks & Invited Lectures

Speaker — Understanding AI Threats to the Chem-Bio-Nuclear World
Roberta Buffett Institute for Global Affairs, Northwestern University · 2026
Workshop on AI-enabled threats and mitigation, presenting the work of the Northwestern Security & AI Lab and its partners. link
Guest Lecture — AI-Based Protein Synthesis: Benefits and Risks
Northwestern University · AI and International Security (265-0-1, Prof. V.S. Subrahmanian) · 2026
Guest Lecture — AI-Based Protein Synthesis: Benefits and Risks
Northwestern University · PHYSICS 101-8 (Prof. V. Kalogera) · 2025
Invited Talk — Demonstration of Generative Malware Models
Conference on AI & National Security, Northwestern University · 2024
link
Guest Lecture — An Introduction to MongoDB
University of Naples Federico II · Big Data Engineering (Prof. V. Moscato) · 2024
Guest Lecture — Overview of Few-Shot Named Entity Recognition
Harvard University · STAT E-100 (Prof. H. Okabe) · 2022

Teaching & Mentorship

Lecturer — Advanced AI Methods and Applications in Healthcare
University of Naples Federico II · PhD Program in ICTH · 2025
Designed and delivered a 10-hour module covering state-of-the-art machine learning techniques, deep learning architectures, and practical applications of AI in clinical and biomedical contexts. flyer
Lecturer — Countering AI Proliferation
Northwestern University · Executive Course · 2025
Co-designed and delivered an executive course with Prof. V.S. Subrahmanian for government and industry stakeholders, covering AI-enabled cyberattacks, IP theft, and malicious use of generative models. link
Teaching Assistant
University of Naples Federico II · 2018–2023
Machine Learning & Big Data for Health, Big Data Engineering, Information Systems, Electronic Calculators I, Elements of Physics I & II, Elements of Informatics.
M.S. Thesis Co-supervisor
University of Naples Federico II · 2020–present
Co-supervised 40+ M.S. theses in machine learning, deep learning, and big data engineering, covering biomedical NLP, explainable AI, knowledge graphs, and recommender systems.
Computational Advisor
Human-Augmented Analytics Group (HAAG), Georgia Tech · 2024–present
Methodological guidance on ML and NLP projects for an interdisciplinary virtual research laboratory.

Research Grants & Funding

Parla Chiaro (Speak Clearly)
Microsoft AI for Good Lab · 2026–present · $48,000 + $20,000 Azure credits
Key Collaborator (PI: Vincenzo Moscato)
Developing the first open-access Italian dialect speech dataset (100+ hours) for healthcare AI safety, and a Dialect-Aware Warning System (DAWS) to protect elderly dialect speakers from AI-generated health misinformation. With cardiologists at Policlinico Federico II and Accenture.
CAP: Countering AI Proliferation
U.S. Department of State · 2025–present · $599,000
Lead Researcher (PI: V.S. Subrahmanian)
Developed a Security Advice Chatbot to support countering the proliferation of AI-enabled threats.
Accelerating Parkinson's Disease Clinical Research in Swallowing and Motor Speech Disorders
American Speech-Language-Hearing Foundation · 2025–present · $10,000
Key Collaborator (PI: Ankita Bhutada)
Coordinated data science and NLP research developing large-language-model solutions to optimize clinical trial efficiency in Parkinson's Disease.
ICARUS: Intelligent Contract Automation for Rethinking User Services
Italian Ministry of Enterprises and Made in Italy · 2022–2024 · $375,825
Scientific Coordinator, Objective 2 (PI: Vincenzo Moscato)
Coordinated research with CNR and Eustema S.p.A. developing AI-based systems for legal document automation, responsible for methodological design, technical supervision, and implementation.

Awards & Honors

2026
Best Paper Award — IEEE CBMS 2026 ↗
39th IEEE International Symposium on Computer-Based Medical Systems — for "Brain3D: Brain Report Automation via Inflated Vision Transformers in 3D."
2026
Editor of Distinction — Editorial Contribution Award ↗
Springer Nature — recognition for top-20% stewardship of the peer-review process as Associate Editor of Artificial Intelligence Review.
2022
Best Definition of Data-Centric AI ↗
Data-Centric AI Community — recognition for an outstanding contribution to defining the emerging field.
2022
Winner, DisTEMIST Challenge ↗
BioASQ — first place in the DISease TExt Mining Shared Task (Team: PICUSLab).
2022
Senior Formative Tutoring Badge ↗
University of Naples Federico II — recognition for excellence in student mentoring and teaching.

Patents

Systems and Methods for Automatic Detection of Human Expression from Multimedia Content
U.S. Patent US20250148826A1 · 2025
A system for analyzing multimedia content featuring a role-matching module to identify participants of interest and a scoring module that evaluates statements based on extracted facial expressions, vocal traits, and textual elements.

Outreach & Media

2026
Featured in IEEE Spectrum for work on the MNW deepfake detection benchmark. "Deepfake Detection Dataset Aims to Keep Up With Generative AI." IEEE Spectrum, May 2026. link
2024–present
Quoted or cited as an expert in AI-generated media detection across international outlets.
  1. "Photos show rescues of U.S. F-15 airmen from Iran in April 2026." PolitiFact, Apr 2026. link
  2. "Imagen muestra que '950 inmigrantes derrumbaron el muro fronterizo entre México y Estados Unidos intentando ingresar a Estados Unidos' en abril 2026." PolitiFact, Apr 2026. link
  3. "A viral audio clip accusing the Clintons in the Epstein case was created with AI." Poynter, Mar 2026. link
  4. "A wave of posts claimed Benjamin Netanyahu was dead. Here's how it spread." Poynter, Mar 2026. link
  5. "Benjamin Netanyahu Is Dead Rumours Explained: Truth Behind 'Cafe Video' as Deepfake Experts Step In." International Business Times, Mar 2026. link
  6. "A narrative that Netanyahu died spread widely on social media. Here's how it unfolded." PolitiFact, Mar 2026. link
  7. "El líder de Corea del Norte Kim Jong Un dijo que 'ningún Dios podrá proteger a Estados Unidos de nuestras nuevas armas.'" PolitiFact, Mar 2026. link
  8. "The Epstein files 'expose Ellen DeGeneres' as a cannibal." PolitiFact, Feb 2026. link
  9. "An audio clip reveals a Jeffrey Epstein victim's account of abuse by the Clintons." PolitiFact, Feb 2026. link
  10. "Video showing PTI's rally in August is likely AI-generated." Soch Fact Check, Aug 2025. link
  11. "JD Vance skewers Musk in viral audio clip. Experts say it's fake." SF Standard, Mar 2025. link
  12. "Audio of Trump calling Imran Khan a convict is a deepfake." Soch Fact Check, Feb 2025. link
  13. "Social media posts misidentify woman burned to death in New York subway fire." USA Today, Jan 2025. link
  14. "A manipulated video of a Pakistani soldier trudging through snow goes viral." Soch Fact Check, Jan 2025. link
  15. "Cek Fakta: Tidak Benar Video yang Diklaim Asteroid Menghantam Bulan." CekFakta, Mar 2025. link
  16. "Video Bulan Dihantam Asteroid, Bagaimana Faktanya?" CekFakta, Mar 2025. link
  17. "No proof Dr. Sebi's cousin was 'sent to life in prison'." PolitiFact, May 2024. link
  18. "Esta llamada al 911 tras caída del puente en Baltimore fue creada con inteligencia artificial." PolitiFact, Apr 2024. link
  19. "An asteroid crashing into the moon? No, video is fabricated." USA Today, Aug 2024. link
  20. "An airplane crashing in California? No, video is altered." USA Today, Aug 2024. link
2022
Featured for research on "Knowledge graphs for next-generation health science applications": "Intelligenza artificiale per la medicina di precisione." Università degli Studi di Napoli Federico II, Mar 2022. link

Service