Polykarpos Meladianos - MBZUAI MBZUAI

Polykarpos Meladianos

Assistant Teaching Professor and Research Scientist

Research Interests

Professor Meladianos' research interests span natural language processing (NLP), large language models (LLMs) and agentic AI systems, retrieval-augmented generation (RAG), graph neural networks and graph mining, real-time summarization and knowledge representation.

Email

Prior to joining MBZUAI, Professor Meladianos led the development of real-world AI systems, including LLM-based chatbots, retrieval-augmented generation pipelines, and fraud detection tools. His current work focuses on agentic architectures, scalable knowledge infrastructures, and real-time AI assistants.

Professor Meladianos is an AI/NLP Engineer with deep industry experience in applied machine learning and natural language processing, having led chatbot and RAG system development using internal LLMs; implemented feedback learning and fine-tuning; and built a multimodal knowledge graph powering RAG. He developed real-time bonus calculation systems and fraud detection using Spark streaming for large-scale ML, created deep learning entity extraction systems for real-time maritime/shipping insights and worked on large-scale text mining, meeting summarization, and deep learning for summarization.
  • Ph.D. in Computer Science, Athens University of Economics and Business
  • Data Science Master-M2 from École Polytechnique
  • Bachelor of Science + M1 in Electrical and Computer Engineering from The Aristotle University of Thessaloniki

  • Graph classification with 2D CNNs, ICANN 2019
  • Kernel Graph Convolutional Neural Networks, ICANN 2018
  • An optimization approach for sub-event detection and summarization in Twitter, ECIR 2018
  • Unsupervised abstractive meeting summarization, ACL 2018
  • A Degeneracy Framework for Graph Similarity, IJCAI 2018
  • Word embeddings from large-scale Greek web content, arXiv 2018
  • K-clique-graphs for dense subgraph discovery, ECML/PKDD 2017
  • Matching node embeddings for graph similarity, AAAI 2017
  • Real-time keyword extraction from conversations, EACL 2017
  • Classifying graphs as images with CNNs, arXiv 2017
  • Combining graph degeneracy and submodularity for extractive summarization, 2017
ORCID

Contact faculty affairs

Interested in working with our renowned faculty?
Fill out the below form and we will get back to you.