David Semedo

Assistant Professor (Tenure Track) · NOVA LINCS · Departamento de Informática, FCT NOVA

prof_pic.jpg

Departament of Informatics

NOVA School of Science and Technology

Universidade NOVA de Lisboa

2829-516 Caparica, Portugal

I am an Assistant Professor (tenure-track) at the Departamento de Informática, FCT — Universidade NOVA de Lisboa, and an integrated researcher at NOVA LINCS. I hold a Ph.D. in Computer Science (FCT NOVA, 2020) and work at the intersection of Vision and Language AI, multimodal deep learning, natural language processing, and computer vision.

My research aims to advance multimodal artificial intelligence systems that can perceive, reason, and act in complex environments — moving beyond pattern recognition towards deliberative planning, situational awareness, and trustworthy behaviour. Current directions include large vision-language models (LVLMs), large language model (LLM) adaptation, conversational and agentic AI, and domain-specific model customisation for Portuguese and European languages.

news

Apr 01, 2026 AMALIA, the national open-source Large Language Model for European Portuguese (PRR programme), was published at PROPOR 2026.
Jul 15, 2025 Our team ranked Runner-up (2nd place) in the Amazon Trusted AI Challenge (2025), as a red team.
Jun 01, 2023 Team TWIZ won 1st place at the Amazon Alexa Prize TaskBot Challenge 2023.

selected publications

  1. AAAI
    FineVAU: A Novel Human-Aligned Benchmark for Fine-Grained Video Anomaly Understanding
    João Pereira, Vasco Lopes, João Neves, and 1 more author
    In Proceedings of the 40th AAAI Conference on Artificial Intelligence, 2026
  2. EACL
    VIGiA: Instructional Video Guidance via Dialogue Reasoning and Retrieval
    Diogo Glória-Silva, David Semedo, and João Magalhães
    In Findings of the Association for Computational Linguistics: EACL 2026, 2026
  3. PROPOR
    AMALIA: A Fully Open Large Language Model for European Portuguese
    Afonso Simplício, Gonçalo Vinagre, Miguel Moura Ramos, and 19 more authors
    In Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026), 2026
  4. EMNLP
    Language Models Can be Efficiently Steered via Minimal Embedding Layer Transformations
    Diogo Tavares, David Semedo, Alexander Rudnicky, and 1 more author
    In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, 2025
  5. EMNLP
    Show and Guide: Instructional-Plan Grounded Vision and Language Model
    Diogo Glória-Silva, David Semedo, and João Magalhães
    In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024
  6. EMNLP
    Multi-trait User Simulation with Adaptive Decoding for Conversational Task Assistants
    Rafael Ferreira, David Semedo, and João Magalhães
    In Findings of the Association for Computational Linguistics: EMNLP 2024, 2024
  7. EACL
    Plan-Grounded Large Language Models for Dual Goal Conversational Settings
    Diogo Glória-Silva, Rafael Ferreira, Diogo Tavares, and 2 more authors
    In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2024), 2024
  8. PROPOR
    GlórIA: A Generative and Open Large Language Model for Portuguese
    Ricardo Lopes, João Magalhães, and David Semedo
    In Proceedings of the 16th International Conference on Computational Processing of Portuguese (PROPOR 2024), 2024
  9. ACM MM
    Understanding News Text and Images Connection with Context-enriched Multimodal Transformers
    Cláudio Bartolomeu, Rui Nóbrega, and David Semedo
    In Proceedings of the 30th ACM International Conference on Multimedia (MM ’22), 2022