Program

AIS-2026-Program


9:00 - 10:40
Giuseppe De Giacomo, University of Oxford, UK

Lecture 1/3 “Foundations of Guardrailing Agentic AI via Temporal Synthesis”
Lecture 2/3 “Foundations of Guardrailing Agentic AI via Temporal Synthesis”


10:40 - 11:20
Coffee break

11:20 - 12:10
Giuseppe De Giacomo, University of Oxford, UK

Lecture 3/3 “Foundations of Guardrailing Agentic AI via Temporal Synthesis”


12:10 - 13:00
Jamie Hayes, Google DeepMind, UK

Lectures 1 “Agentic Security (how to attack and defend against prompt injection)”


13:00 - 15:00
Lunch

15:00 - 15:50
Jamie Hayes, Google DeepMind, UK

Lectures 2 “Agentic Security (how to attack and defend against prompt injection)”


15:50 - 16:40
Christoph Feichtenhofer, Meta Superintelligence Labs, USA

Lectures 1 “On the impact of data for vision-language learning”


16:40 - 17:20
Coffee break

17:20 - 19:00
Christoph Feichtenhofer, Meta Superintelligence Labs, USA

Lectures 2 “On the impact of data for vision-language learning”
Lecture 3/3 “Building frontier visual segmentation models”


19:00 - 20:00
Free Time

20:00
Dinner

9:00 - 10:40
Rafal Bogacz, University of Oxford, UK

Lecture 1/2 “Predictive coding: training neural networks with local rules”
Lecture 2/2 “Predictive coding: advantages over backpropagation”


10:40 - 11:20
Coffee break

11:20 - 13:00
Sebastian Lins Risi, Sakana AI, Japan & IT University of Copenhagen, Denmark

Lecture 1/3 “Neuroevolution: The basics”
Lecture 2/3 “Neuroevolution: Diversity, Novelty, and Open-Ended Discovery”


13:00 - 15:00
Lunch

15:00 - 15:50
Sebastian Lins Risi, Sakana AI, Japan & IT University of Copenhagen, Denmark

Lecture 3/3 “Neuroevolution: Synergies with Generative AI”


15:50 - 16:40
Sumi Helal, University of Bologna, Italy & University of Florida, USA

Lecture “Generative Smart Spaces – A Novel Physical-AI Approach to Programming Future Internet-of-Things”


16:40 - 17:20
Coffee break

17:20 - 20:00
Oral Session 1
The Crown Conference Room
  • 17:20 – 17:35
    Xiao Jin, Conner Parker and Rui Feng,
    “Structure-Informed Multi-Head Attention for Correlated Predictors”
  • 17:35 – 17:50
    Evann Rabeau, Jérôme Iste, Ismael Freire and Mehdi Khamassi,
    “Model-free multi-agent reinforcement learning simulation of the influence of empathy on the sustainability of agents’ consumption in a common-pool resource dilemma”
  • 17:50 – 18:05
    Tian Li, Huizhi Liang, Lucy Robinson and Varun Ojha,
    “Predicting Cognitive Behavioural Therapy Outcomes based on Clinical Assessments and Referral Letters”
  • 18:05 – 18:20
    “Valerio Bellandi and Stefano Siccardi,
    “Predicting Energy Consumption in Buildings and Mobility with ML and Graphs”
  • 18:20 – 18:35
    Ilinca Tiriblecea and Gabriel Turinici,
    “Inferring Missing Trajectory Data with Temporal Convolutional Networks”
  • 18:35 – 18:50
    Qingliang Meng, Luogeng Xiong, Wei Liang, Huizhi Liang and Tian Li,
    “Unified Timbre-Style Representation Learning for Controllable Speech Synthesis”
  • 18:50 – 19:05
    Simon Hakenes and Tobias Glasmachers,
    “Macro-Action Topological Navigation under Noisy Localization using Reinforcement Learning”
  • 19:05 – 19:20
    Jianfei Xu, Huizhi Liang, Judith Harrison, John-Paul Taylor, Jaume Bacardit and Varun Ojha,
    “HypTI: Hyperbolic Geometry-Aware Multimodal Fusion Framework for Dementia Diagnosis Based on Tabular-Image Data”
  • 19:20 – 19:35
    Tõnn Talpsepp, Kristjan Liivamägi and Tarvo Vaarmets,
    “What Does a Transformer Learn from an Investor’s Trade Sequence and Educational Profile?”
  • 19:35 – 19:50
    Mark Callan and Meghana Kshirsagar,
    “Grounded Multi-Agent Reasoning for Financial Due Diligence: A Deterministic-First Architecture with Claim- Level Verification”

20:00
Social Dinner

9:00 - 10:40
Oral Session 2
The Da Vinci Conference Room

(remote presentations)

  • 09:00 – 09:15
    Giuseppe Trimigno and Gianfranco Lombardo,
    “Prompt-Induced Schema Awareness for Question Answering over Knowledge Graphs”
  • 09:15 – 09:30
    Gabriel Turinici,
    “Softmax gradient policy for variance minimization and risk-averse multi armed bandits”
  • 09:30 – 09:45
    Muhammad Hashim, Roberto Rugani, and Marco Raugi,
    “Techno-Economic and Microscopic Simulation of DC Fast Charging for Electric Vehicles in Renewable Energy Communities”
  • 09:45 – 10:00
    Nadejda Bocheva, Miroslava Stefanova and Milena Mihaylova,
    “Drift-Diffusion Modeling of Orientation Effects in Contour-Defined Triangle Detection under Time Pressure”
  • 10:00 – 10:15
    Sarah Youssef, Hussein Assaf, Esam Abdel-Raheem and Ziad Kobti,
    “Clinical Prior-Guided Multimodal Learning Framework for Brain Tumor Segmentation”
  • 10:15 – 10:30
    Jose Barrera-Garcia, Victor Macho, Broderick Crawford, Álvaro Olmedo, Felipe Cisternas-Caneo, Yoslandy Lazo,
    Ricardo Soto, José Maria Jurado Carmona, Giovanni Giachetti and José Manuel Gómez-Pulido,
    “When Features Leak: A Leakage Audit and Imbalance Benchmark for Manchester Triage”
  • 10:30 – 10:45
    Nicola Floris, Gianmarco Cherchi, Gianni Fenu, Marco Manolo Manca, Roberto Saia, Riccardo Scateni and
    Alessandro Sebastian Podda,
    “A Data-Centric, Comparative Approach to Real-Time Pedestrian Orientation Estimation for Autonomous Driving”

10:40 - 11:20
Coffee break

11:20 - 13:00
Oral Session 3
The Da Vinci Conference Room

(remote presentations)

  • 11:20 – 11:35
    Asma Atamna, Tom Maus, Fabian Kievelitz and Tobias Glasmachers,
    “Cumulative Learning Rate Adaptation: Revisiting Path-Based Schedules for SGD and Adam”
  • 11:35 – 11:50
    Marco Pilloni, Alessandro Giuliani, Alessandro Sebastian Podda, Alessia Pisu, Livio Pompianu and Salvatore Carta,
    “Semantic Normalization of Coercive Knowledge Graph Triples with Large Language Models”
  • 11:50 – 12:05
    Gauri Vaidya, Gauri Mehta, Yuanyuan Chen, Alison O’Connor and Meghana Kshirsagar,
    “A Static Digital Twin Framework with Predictive Analytics for Healthcare Operational Intelligence”
  • 12:05 – 12:20
    Minhquang Nguyen, Vanha Tran, Thiloan Bui, Truongminh Ngo and Thanhtu Nguyen,
    “Domain-Adapted Embedding Fine-tuning based ESG Question-Answering System Using GraphRAG and Large Language Models: A Case Study for the Vietnamese Banking Sector”
  • 12:20 – 12:35
    Melchiorre Cosseddu, Salvatore Carta, Alessandro Sebastian Podda, Livio Pompianu, Marco Manolo Manca and Alessandro Giuliani,
    “Mapping Business Ecosystems with LLMs: An Automated Pipeline for Knowledge Graph Construction from Corporate Websites”
  • 12:35 – 12:50
    José Luis Carrillo-Medina, Ivone Arias-Almeida, Pablo Morales-Paredes and Kevin Chuquitarco-Velasco,
    “A Two-Stage Framework for Facial Identification and Torso-Centered Person Re-Identification Using Body Silhouette and Clothing Color in Non-Overlapping Cameras”
  • 12:50 – 13:05
    Cheng Arn David Goh,
    “Signed Decomposition of the Regression Structure via Linear JEPA Training: An Algorithm, Empirical Validation, and a Machine-Checked Proof”

13:00 - 15:00
Lunch

15:00 - 16:40
Oral Session 4
The Da Vinci Conference Room

(remote presentations)

  • 15:00 – 15:15
    Shasha Yu, Fiona Carroll and Barry L. Bentley,
    “Execution-Refusal Profiling of Tool-Augmented LLM Agents: Behavioural Characterization Across Autonomy Scaffolds”
  • 15:15 – 15:30
    Elena Bellodi and Francesco Resca,
    “Smart City Analytics for Sustainable Tourism: an AI-Driven Framework for Flow Forecasting in a Historic Borough”
  • 15:30 – 15:45
    Elena Bellodi and Francesco Resca,
    “Forecasting and Interpreting Urban Air Pollution Under Weather, Traffic and Airport Influence”
  • 15:45 – 16:00
    Emmanuel Osei-Brefo, Huizhy Liang, Carlos Amador, Gustavo Leon Cazares, Matthew Davis and Varun Ojha,
    “GenAnalysis: A Contract-Driven Multi-Agent Framework for Auditable Data Analysis and Interpretation”
  • 16:00 – 16:15
    Maryam Mosleh and Ellis Solaiman,
    “Faithful Role-Aware Explanations for Adaptive Learning: Translating Structured XAI Evidence with LLMs”
  • 16:15 – 16:30
    Shu Li, Huizhi Liang, Wanqing Zhao, Mohamad Kassem, Ankush Prashar and Varun Ojha,
    “HypCR: Hyperbolic Convolution Recurrent Model with Mixture-of-Experts for Multivariate Time-Series Forecasting”

16:40 - 19:00
Marine Le Morvan, INRIA Saclay, France

Lecture 1/3 “The rise of Tabular Foundation Models”
Lecture 2/3 “TabICL: on open Tabular Foundation Model”
Lecture 3/3 “Next frontiers for TFMs”


19:00 - 20:00
Oral Session 5
The Crown Conference Room
  • 19:00 – 19:15
    Andrea de Ruvo, Pierluigi Castelli, Nicolas Radomski and Adriano Di Pasquale,
    “PhenoCal: Natural-Control Label Calibration for Cross-Cohort Machine Learning”
  • 19:15 – 19:30
    Stacey Truex and Margaret Malan,
    “Miles: A System for Parameter Analysis in Practical, Private Federated Learning”
  • 19:30 – 19:45
    Yannick Kees, Elena Hoemann, Frank Köster and Sven Hallerbach,
    “Holistic Risk Assessment for AI Systems”
  • 19:45 – 20:00
    Francesco Morano and Giovanni Denaro,
    “Phishing4training: an approach to train users on highly personalised phishing”

20:00
Dinner

9:00 - 9:50
George Paliouras, National Centre for Scientific Research “Demokritos”, Athens, Greece

Lecture “Neuro-Symbolic Reasoning and Learning for Complex Event Recognition/Forecasting”


9:50 - 10:40
Maria K. Eckstein, Google DeepMind, London, UK

Lecture 1/3 “Computational Cognitive Modeling Using Reinforcement Learning”


10:40 - 11:20
Coffee break

11:20 - 13:00
Maria K. Eckstein, Google DeepMind, London, UK

Lecture 2/3 “Advancing Cognitive Modeling by Combining Hand-Crafted and Neural-Network Models”
Lecture 3/3 “Advancing Cognitive Modeling using Procedurally-Generated Task Spaces”


13:00 - 15:00
Lunch

15:00 - 20:00
Oral Session 6
The Da Vinci Conference Room
  • 15:00 – 15:15
    Ioannis Demetriou and Eva Perivolari,
    “Explainable and Monotonic Machine Learning for Hotel Booking Cancellation Prediction”
  • 15:15 – 15:30
    Derrick Chan and Dustin van der Haar,
    “Deep Learning Approaches to Restore Rainy Images in Adverse Weather Conditions”
  • 15:30 – 15:45
    Ioannis Sfyrakis, Adam Booth, Rajiv Ranjan and Ellis Solaiman,
    “EdgeCross: A Distributed AI-Assisted Framework for DeFi Smart Contract Auditing”
  • 15:45 – 16:00
    Absalom Ezugwu,
    “A Unified World Model for Predictive Representation Learning and Latent-Space Decision-Making”
  • 16:00 – 16:15
    Michal Skalican, Patrik Kozlik, Filip Zatroch, Maximilian Strecansky, Stefan Dlugolinsky, Martin Bobak, Ladislav Hluchy and Igor Skrlec,
    “Explainable AI-Driven Pipeline Composition for Cloud-Edge Data Platforms”
  • 16:15 – 16:30
    Tahitoa Leygue, Astrid Sabourin, Christian Bolzmacher, Sylvain Bouchigny, Margarita Anastassova and Quoc-Cuong Pham,
    “CADENCE: Congruence-Aware Word-Level Fusion for Speech Emotion Recognition”
  • 16:30 – 16:45
    Agate Jarmakovica and Arnis Kirshners,
    “Transformer-Based Semantic Variable Layer for AI-Assisted Quality Assessment of Tabular Healthcare Data”
  • 16:45 – 16:55
    Benny Stein and Abhinav Pareek,
    “A Trace-Based Monitoring Approach for Trustworthy Agentic AI”
    (Oral Presentation)

(remote presentations)

  • 16:55 – 17:10
    Nikolay Georgiev,
    “Systemic Divergence in LLM-Based Personality Trait Estimation from Text Poses Risks for Psychological Applications”
    (Oral Presentation)
  • 17:10 – 17:25
    Kexin Zheng and T. K. Satish Kumar,
    “A FastMap-Based Framework for Black-Box Optimization”
    (Oral Presentation)
  • 17:25 – 17:40
    Vernon Kok, Absalom Ezugwu and Michael Olusanya,
    “Adaptive Object-Aware Mapless Mobile Robot Navigation Using YOLOE and Soft Actor-Critic”
    (Oral Presentation)
  • 17:40 – 17:55
    Jorge Jabón Domínguez,
    “MGGAN: Learning a Multiscale Graph-based GAN for non-parametric 3D single-shape deformation and variation”
    (Oral Presentation)
  • 17:55 – 18:05
    Rommel Regis,
    “Prompt Engineering for Learning Mathematical Optimization with ChatGPT”
    (Education Track)
  • 18:05 – 18:15
    Vera Kalinichenko, “Vibe Coded Applications in the Industry Settings”
    (Oral Presentation)
  • 18:15 – 18:25
    Mireia Louzan Calo, Sergio Consoli, Alessandro Zani, Mario Ceresa and Vicenç Gómez,
    “Federated Learning for Multimodal Wearable Human Activity Recognition: Preliminary Experiments”
    (Short Paper)
  • 18:25 – 18:40
    Hasibul Karim Shanto and R. M. Shahbab,
    “Towards Robust and Stable Single Decision Trees under Distributional Shift”
    (Late Breaking Papers)
  • 18:40 – 18:55
    Alejandro Aybar Flores, Rocío Maehara and Luis Benites,
    “Explaining and Segmenting Inequality Perceptions in Peru: Regression Evidence and Clustering from ENADES 2024” (Late Breaking Papers)
  • 18:55 – 19:05
    Mohamed El Moutaouadia El Idrissi, Mohamed Sraitih, Parisa Ghodous, Guillaume Besch, Romain Bronnert and Amir Hajjam El Hassani,
    “Safe Active Learning for Patient-Specific Pharmacodynamic Identification in Propofol Sedation”
    (Short Paper)
  • 19:05 – 19:15
    Pankaj Kumar Chauhan, Sachchida Nand Chaurasia and Shyam Sundar,
    “A Reinforcement Learning based Evolutionary Hyperheuristic for the Coloured Balanced TSP”
    (Oral Presentation)

 


20:00
Dinner