Dr. Yashar Deldjoo

Senior Researcher & Assistant Professor, Polytechnic University of Bari

Dr. Yashar Deldjoo is a senior research scientist and a tenure-track Assistant Professor in Computer Science at the Polytechnic University of Bari, Italy, recognized among the top 2% of researchers worldwide (Stanford’s Ioannidis database, 2023–2024). His work focuses on integrating elements of trustworthy and responsible AI (e.g., fairness, adversarial robustness, security) into recommender systems. His lab’s current efforts address the potential hazards of large language models (LLMs)—such as hallucinations, randomness, and bias—in RecSys and downstream ML tasks, developing novel evaluation metrics to ensure robust, trustworthy AI.

He recently led a multidisciplinary collaboration involving Amazon, DeepMind, and several leading universities for an upcoming book at FN&TIR'25, Next-Generation Recommender Systems under Generative AI (Gen-RecSys), which explores advanced generative techniques. A widely recognized survey version of this book was presented at KDD'24. He also discussed next-generation Gen-RecSys on the Recsperts Podcast. Another line of his research concerns multimodal learning for RecSys, Fashion AI, and the music industry. He serves as Associate Editor for IEEE TKDE and ACM CSUR, Distinguished Reviewer for ACM TORS, and Guest Editor for special issues on Generative RecSys (2025) and Trustworthy RecSys (2024). He is a Senior Program Committee (SPC) member at SIGIR, CIKM, ECAI, TheWebConf, and ACM RecSys.

Highlights

Forthcoming Book: “Recommendation with Generative Models” (2025)
Workshop: Co-organizing ROEGEN @ ACM RecSys 2024 (Risks & Opportunities of Generative Models)
Podcast: Generative RecSys on Recsperts

Research

I investigate reliable and trustworthy AI in Recommender Systems, emphasizing:

Adversarial robustness & attack defense
Fairness & bias mitigation
Explainability & interpretability
Generative AI & large language models

My aim is to bridge theoretical models and real-world deployment, ensuring AI is socially responsible.

News

Joining IEEE TKDE: Appointed as Associate Editor at IEEE Transactions on Knowledge and Data Engineering (TKDE).
Special Issue on Generative RecSys: We are co-editing a special issue at ACM TORS (2025) —call for papers now open.

Note to Students

I'm interested in collaborating with students/post-docs in:

Generative Recommender Systems
Ethics & Trust in AI/ML
Evaluation Metrics for Risk/Hazards

Feel free to reach out if you are passionate about pushing AI innovation responsibly.

Editorial & Service Roles

Associate Editor: IEEE TKDE, ACM Computing Surveys (CSUR)
Distinguished Reviewer: ACM Transactions on Recommender Systems (TORS)
Guest Editor: TORS Special Issues on Trustworthy RecSys (2024) & Generative RecSys (2025)
Senior Program Committee: SIGIR, CIKM, ECAI, TheWebConf, ACM RecSys

Publications

A thematic selection of my most impactful works is below. For the complete list, see Google Scholar or DBLP .

Generative Models for RecSys

Works on VAE, Diffusion Models, GANs, LLMs, and multimodal approaches for next-generation recommendations.

Recommendation with Generative Models (Book, 2025)
Y. Deldjoo, M. S. Mansour, M. Larson, G. Faggioli
Venue: Publisher TBA
A Review of Modern Recommender Systems using Generative Models (Gen-RecSys) (Under Review, 2024)
Y. Deldjoo, L. Belli, P. Cremonesi, T. Di Noia
Venue: Under Review

Trustworthy Gen-RecSys (Biases & Security)

Fairness, bias mitigation, and adversarial tactics in LLM-driven recommenders.

Understanding Biases in ChatGPT-based Recommender Systems (2024)
Y. Deldjoo, V. W. Anelli, T. Di Noia
Venue: Under Review
CFaiRLLM: Consumer Fairness Evaluation in LLM-based Recommender Systems (2024)
Y. Deldjoo, T. Di Noia
Venue: Under Review
Poison-RAG: Adversarial Data Poisoning Attacks on Retrieval-Augmented Generation (2024)
Y. Deldjoo, F. A. Merra, T. Di Noia
Venue: Under Review

Trustworthy in Classical RecSys

Fairness, bias mitigation, and adversarial defenses for traditional recommendation approaches.

Cpfair: Personalized Consumer and Producer Fairness Re-ranking (ACM TORS, 2024)
H. A. Rahmani, M. Naghiaei, Y. Deldjoo
Fairness in Recommender Systems: Research Landscape and Future Directions (UMUAI, 2023)
Y. Deldjoo, T. Di Noia, F. A. Merra, M. S. Mansour
A Survey on Adversarial Recommender Systems (ACM CSUR, 2022)
Y. Deldjoo, T. Di Noia, F. A. Merra

Data & Dataset Characteristics

How dataset properties influence model vulnerability, fairness, and overall performance.

How Dataset Characteristics Affect the Robustness of Collaborative Recommendation Models (2023)
Y. Deldjoo, V. W. Anelli, H. A. Rahmani, M. Naghiaei
Venue: Under Review
Explaining Recommender Systems Fairness and Accuracy through the Lens of Data (2023)
Y. Deldjoo, V. W. Anelli, M. Naghiaei
Venue: Under Review

Evaluation Frameworks (Fairness, Gen-RecSys)

Unified frameworks and metrics to systematically measure fairness and generative model performance.

A Flexible Framework for Evaluating User and Item Fairness in Recommender Systems (2023)
Y. Deldjoo, L. Belli, G. Faggioli, T. Di Noia
Venue: Under Review
A Unifying and General Account of Fairness Measurement in Recommender Systems (2023)
Y. Deldjoo, T. Chen, N. Mansour, T. Di Noia
Venue: Under Review

Application of LLM in Downstream RecSys/ML Tasks

Leveraging large language models for specialized domains such as healthcare.

ChatGPT-HealthPrompt: Harnessing XAI in Prompt-Based Healthcare Decision Support (XI-ML@ECAI, 2023)
F. Nazary, Y. Deldjoo, T. Di Noia

Multimodal RecSys & Fashion AI

Integrating visual, textual, and conversational signals in domains like fashion and beyond.

Towards Multi-Modal Conversational Information Seeking (2023)
Y. Deldjoo, V. W. Anelli, P. Lops, T. Di Noia
Venue: Under Review
A Review of Modern Fashion Recommender Systems (ACM CSUR, 2023)
Y. Deldjoo, D. Malitesta, T. Chen, F. A. Merra

Surveys

Fairness in Recommender Systems: Research Landscape and Future Directions (UMUAI, 2023)
Y. Deldjoo, T. Di Noia, F. A. Merra, M. S. Mansour
[Link]
A Survey on Adversarial Recommender Systems (ACM CSUR, 2022)
Y. Deldjoo, T. Di Noia, F. A. Merra
[Link]

Tutorials

Understanding Language Modeling Paradigm Adaptations in Recommender Systems (ECAI'24)
L. Zhang, P. Liu, Y. Deldjoo, A. Felfernig
Pursuing Privacy in Recommender Systems (RecSys'21)
V. W. Anelli, L. Belli, Y. Deldjoo, T. Di Noia

Book & Book Chapters

Recommendation with Generative Models (Forthcoming, 2025)
Y. Deldjoo, M. S. Mansour, M. Larson, G. Faggioli
Multimedia Recommender Systems: Algorithms and Challenges (Recommender Systems Handbook, 2023)
Y. Deldjoo, V. W. Anelli, M. Schedl
Adversarial Recommender Systems: Attack, Defense, and Advances (Recommender Systems Handbook, 2023)
V. W. Anelli, Y. Deldjoo, L. Belli

Other Technical Papers

Exploring the Impact of Temporal Bias in Point-of-Interest Recommendation (RecSys, 2022)
H. A. Rahmani, M. Naghiaei, A. Tourani, Y. Deldjoo
Audio-Visual Encoding of User Facial Expressions for Emotion-Aware Recommendations (RecSys, 2018)
Y. Deldjoo, M. Schedl, P. Lops, T. Di Noia

Professional Activities

Aug 2022 – Present: Tenure-track Assistant Professor, Polytechnic University of Bari
Aug 2019 – Jul 2022: Assistant Professor (RtD-A), Polytechnic University of Bari
Apr 2019 – Jul 2019: Postdoctoral Researcher, Polytechnic University of Bari
Jan 2018 – Feb 2019: Research Assistant, University of Milano-Bicocca

Education

Ph.D. in Computer Science, Politecnico di Milano (2014–2018)
M.Sc. in Electrical Engineering, Chalmers University of Technology (2008–2010)
B.A. in English-Linguistics, University of Gothenburg (2007–2011)
B.Sc. in Electrical Engineering, University of Guilan (2002–2006)