Hi!

I am a tenure-track assistant professor at CIMeC and DISI, University of Trento (Italy). I spend my time pondering open problems in AI and Machine Learning and coming up with practical (and sometimes sound) solutions for them. I have previously worked as a post-doc with Andrea Passerini and Fausto Giunchiglia at DISI, with Luc De Raedt at KU Leuven (Belgium), and with Luciano Serafini at FBK (Trento, Italy). I obtained my Ph.D. from the University of Trento in 2014.

On Google Scholar, my pen name is “uaQCyXkAAAAJ”. All the code I write is available on GitHub. Sometimes I tweet and retweet AI and ML-related content @looselycorrect.

I maintain a curated list of literature on interacting with Machine Learning models via explanations in the awesome explanatory supervision repo. Contributions are welcome!

Contacting Me

Feel free to contact me at: name.surname@unitn.it

Research Topics

I am interested in AI and Machine Learning, and specifically in ensuring that models learned from data are aligned with the specifications, requirements, and preferences of their users, also out-of-distribution, in interactive and lifelong settings. I am attacking this problem through a combination of:

  • xai Explainable Artificial Intelligence,
  • cbms Self-Explainable / Concept-Based Models,
  • nesy Neuro-Symbolic Integration / Statistical Relational Learning,
  • iml Interactive Machine Learning,
  • llm Large Language Models / Foundation Models,
  • cl Constraints and Constraint Learning,
  • pe Preference Elicitation.

If you are a student and you are interested in these topics, have a look at the theses and projects page.

Publications & Preprints

2024

  • Learning To Guide Human Decision Makers With Vision-Language Models. Debodeep Banerjee, Stefano Teso, Burcu Sayin, Andrea Passerini. Under submission. [preprint] xai llm

  • Towards Logically Consistent Language Models via Probabilistic Reasoning. Diego Calanzone, Stefano Teso, Antonio Vergari. ICLR 2024 Workshop on Reliable and Responsible Foundation Models. [preprint] nesy llm

  • BEARS Make Neuro-Symbolic Models Aware of their Reasoning Shortcuts. Emanuele Marconato, Samuele Bortolotti, Emile van Krieken, Antonio Vergari, Andrea Passerini, Stefano Teso. Spotlight at UAI 2024 [preprint, code] nesy

2023

  • Not all neuro-symbolic concepts are created equal: Analysis and mitigation of reasoning shortcuts. Emanuele Marconato, Stefano Teso, Antonio Vergari, Andrea Passerini. NeurIPS 2023. [paper, code] nesy

  • Interpretability Is in the Mind of the Beholder: A Causal Framework for Human-Interpretable Representation Learning Emanuele Marconato, Andrea Passerini, Stefano Teso. Entropy. [paper, preprint] iml xai cbms

  • How Faithful are Self-Explainable GNNs? Marc Christiansen, Lea Villadsen, Zhiqiang Zhong, Stefano Teso, Davide Mottin. LOG 2023. [paper, code] xai

  • Learning to Guide Human Experts via Personalized Large Language Models. Debodeep Banerjee, Stefano Teso, Andrea Passerini. Workshop @ Conference. [paper] iml xai

  • To Transfer or Not to Transfer and Why? Meta-Transfer Learning for Explainable and Controllable Cross-Individual Activity Recognition. Qiang Shen, Stefano Teso, Fausto Giunchiglia, Hao Xu. Electronics. [paper] other

  • Neuro-Symbolic Reasoning Shortcuts: Mitigation Strategies and their Limitations. Emanuele Marconato, Stefano Teso, Andrea Passerini. NeSy Workshop 2023. [paper] nesy

  • Neuro Symbolic Continual Learning: Knowledge, Reasoning Shortcuts and Concept Rehearsal. Emanuele Marconato, Gianpaolo Bontempo, Elisa Ficarra, Simone Calderara, Andrea Passerini, Stefano Teso. ICML 2023. [paper], code] nesy

  • Semantic Loss Functions for Neuro-Symbolic Structured Prediction. Kareem Ahmed, Stefano Teso, Paolo Morettin, Luca Di Liello, Pierfrancesco Ardino, Jacopo Gobbi, Yitao Liang, Eric Wang, Kai-Wei Chang, Andrea Passerini, Guy Van den Broeck. Compendium of Neurosymbolic Artificial Intelligence. [book chapter] nesy

  • Personalized Algorithmic Recourse with Preference Elicitation. Giovanni De Toni, Paolo Viappiani, Stefano Teso, Bruno Lepri, Andrea Passerini. TMLR. [paper, code] iml xai

  • Leveraging Explanations in Interactive Machine Learning: An Overview. Stefano Teso, Öznur Alkan, Wolfgang Stammer, Elizabeth Daly. Frontiers in Artificial Intelligence. [paper, preprint] iml xai

2022

  • Machine Learning for Combinatorial Optimisation of Partially-Specified Problems: Regret Minimisation as a Unifying Lens. Stefano Teso, Laurens Bliek, Andrea Borghesi, Michele Lombardi, Neil Yorke-Smith, Tias Guns, Andrea Passerini. [preprint] cl

  • Concept-level Debugging of Part-Prototype Networks. Andrea Bontempelli, Stefano Teso, Fausto Giunchiglia, Andrea Passerini. [preprint, code] iml xai

  • Learning Max-SAT from Contextual Examples for Combinatorial Optimisation. Mohit Kumar, Samuel Kolb, Stefano Teso, Luc De Raedt. Artificial Intelligence. [paper] cl

  • Human-in-the-loop Handling of Knowledge Drift. Andrea Bontempelli, Fausto Giunchiglia, Andrea Passerini, Stefano Teso. Data Mining and Knowledge Discovery. [pdf, code] iml

  • GlanceNets: Interpretabile, Leak-proof Concept-based Models. Emanuele Marconato, Andrea Passerini, Stefano Teso. NeurIPS 2022. [pdf, code] cbms xai

  • Semantic Probabilistic Layers for Neuro-Symbolic Learning. Kareem Ahmed, Stefano Teso, Kai-Wei Chang, Guy Van den Broeck, Antonio Vergari. NeurIPS 2022. [pdf] nesy

  • Machine Learning for Utility Prediction in Argument-Based Computational Persuasion. Ivan Donadello, Anthony Hunter, Stefano Teso, Mauro Dragoni. AAAI 2022. [pdf] other

  • Federated Multi-Task Attention for Cross-Individual Human Activity Recognition. Qiang Shen, Haotian Feng, Rui Song, Stefano Teso, Fausto Giunchiglia, Hao Xu. IJCAI 2022. [pdf] other

  • Catastrophic Forgetting in Continual Concept Bottleneck Models. Emanuele Marconato, Gianpaolo Bontempo, Stefano Teso, Elisa Ficarra, Simone Calderara, Andrea Passerini. ICIAP 2021. [paper] cbms

  • Toward a Unified Framework for Debugging Concept-based Models. Andrea Bontempelli, Fausto Giunchiglia, Andrea Passerini, Stefano Teso. Workshop on Interactive Machine Learning @ AAAI 2022. [pdf] iml xai cbms

  • Efficient and Reliable Probabilistic Interactive Learning with Structured Outputs. Stefano Teso, Antonio Vergari. Workshop on Interactive Machine Learning @ AAAI 2022. [pdf] iml nesy

  • Lifelong Personal Context Recognition. Andrea Bontempelli, Marcelo Rodas Britez, Xiaoyue Li, Haonan Zhao, Luca Erculiani, Stefano Teso, Andrea Passerini, Fausto Giunchiglia. Workshop on Human-Centered Design of Symbiotic Artificial Intelligence @ HHAI 2022. [pdf] iml nesy

2021

  • Bandits for Learning to Explain from Explanations Frey Behrens, Stefano Teso, Davide Mottin. [preprint, code] iml xai

  • Learning Modulo Theories for Constructive Preference Elicitation. Paolo Campigotto, Stefano Teso, Roberto Battiti, Andrea Passerini. Artificial Intelligence. [paper] pe cl

  • Putting Human Behavior Predictability in Context. Wanyi Zhang, Qiang Shen, Stefano Teso, Bruno Lepri, Andrea Passerini, Ivano Bison, Fausto Giunchiglia. EPJ Data Science. [pdf] other

  • Interactive Label Cleaning with Example-based Explanations. Stefano Teso, Andrea Bontempelli, Fausto Giunchiglia, Andrea Passerini. NeurIPS 2021. [pdf, code] iml xai

  • A Compositional Atlas of Tractable Circuit Operations: From Simple Transformations to Complex Information-Theoretic Queries. Antonio Vergari, YooJung Choi, Anji Liu, Stefano Teso, Guy Van den Broeck. NeurIPS 2021. [pdf] srl

  • Learning Mixed-Integer Linear Programs from Contextual Examples. Mohit Kumar, Stefano Kolb, Luc De Raedt, Stefano Teso. Data Science Meets Optimization Workshop @ IJCAI 2021. [pdf] cl

  • Neuro-Symbolic Constraint Programming for Structured Prediction. Paolo Dragone, Stefano Teso, Andrea Passerini. Workshop on Neural-Symbolic Learning and Reasoning @ IJCLR 2021. [pdf] nesy

  • Co-creating Platformer Levels with Constrained Adversarial Networks. Paolo Morettin, Andrea Passerini, Stefano Teso. Workshop on Human-AI Co-Creation with Generative Models @ IUI 2021. [pdf] nesy

2020

  • Machine Guides, Human Supervises: Interactive Learning with Global Explanations. Teodora Popordanoska, Mohit Kumar, Stefano Teso. [preprint] iml xai

  • Making Deep Neural Networks Right for the Right Scientific Reasons by Interacting with their Explanations Patrick Schramowski, Wolfgang Stammer, Stefano Teso, Anna Brugger, Franziska Herbert, Xiaoting Shao, Hans-Georg Luigs, Anne-Katrin Mahlein, Kristian Kersting. Nature Machine Intelligence. [paper] iml xai

  • Predictive Spreadsheet Autocompletion with Constraints Samuel Kolb, Stefano Teso, Anton Dries, Luc De Raedt. Machine Learning. [pdf] srl cl

  • Challenges in Interactive Machine Learning. Stefano Teso, Oliver Hinz. Künstliche Intelligenz. [pdf] iml

  • Learning Weighted Model Integration Distributions. Paolo Morettin, Samuel Kolb, Stefano Teso, Andrea Passerini. AAAI 2020. [pdf] srl cl

  • Learning Max-SAT from contextual examples for combinatorial optimisation. Mohit Kumar, Samuel Kolb, Stefano Teso, Luc De Raedt. AAAI 2020 [pdf] cl

  • Efficient Generation of Structured Objects with Constrained Adversarial Networks Luca Di Liello, Piero Ardino, Jacopo Gobbi, Paolo Morettin, Stefano Teso, Andrea Passerini. NeurIPS 2020. [pdf] nesy

  • Learning in the Wild with Incremental Skeptical Gaussian Processes. Andrea Bontempelli, Stefano Teso, F Giunchiglia, Andrea Passerini. IJCAI 2020 [pdf, code] iml

  • Human-Machine Collaboration for Democratizing Data Science. Clement Gautrais, Yann Dauxais, Stefano Teso, Samuel Kolb, Gust Verbruggen, Luc De Raedt. Human-Like Machine Intelligence. [pdf] cl

  • VisualSynth: Democratizing Data Science in Spreadsheets. Clement Gautrais, Yann Dauxais, Samuel Kolb, Architt Jain, Mohit Kumar, Stefano Teso, Elia Van Wolputte, Gust Verbruggen, Luc De Raedt. ECML-PKDD 2020 Applied Data Science and Demo Track. [paper] cl

  • Toward Machine-Guided, Human-Initiated Explanatory Interactive Learning. Teodora Popordanoska, Mohit Kumar, Stefano Teso. TAILOR Workshop @ ECAI 2020. [pdf] iml xai

  • Does Symbolic Knowledge Prevent Adversarial Fooling? Stefano Teso. Workshop on Statistical Relational AI @ NeurIPS 2020. [pdf] nesy

  • Multi-Modal Subjective Context Modelling and Recognition. Qiang Shen, Stefano Teso, Wanyi Zhang, Hao Xu, Fausto Giunchiglia. Workshop on Modelling and Reasoning in Context @ ECAI 2020. [pdf] other

2019

  • Combining learning and constraints for genome-wide protein annotation Stefano Teso, Luca Masera, Michelangeli Diligenti, Andrea Passerini. BMC Bioinformatics. [paper] nesy

  • Constraint Learning: An Appetizer. Stefano Teso. Reasoning Web. Explainable Artificial Intelligence. [paper] cl

  • Explanatory Interactive Machine Learning. Stefano Teso, Kristian Kersting. AIES 2019 [pdf, code] iml xai

  • Acquiring Integer Programs from Data. Mohit Kumar, Stefano Teso, Luc De Raedt. IJCAI 2019. [pdf] cl

  • Learning Linear Programs from Data. Elias Arnold Schede, Samuel Kolb, Stefano Teso. ICTAI 2019. [paper] cl

  • Automating Personnel Rostering by Learning Constraints Using Tensors. Mohit Kumar, Stefano Teso, P De Causmaecker, Luc De Raedt. ICTAI 2019. [paper] cl

  • SynthLog: A Language for Synthesising Inductive Data Models. Yann Dauxais, Clement Gautrais, Anton Dries, Architt Jain, Samuel Kolb, Mohit Kumar, Stefano Teso, Elia Van Wolputte, Gust Verbruggen, Luc De Raedt. ECML-PKDD 2019. [paper] cl

  • Acquiring Non-Linear Constraints. Mohit Kumar, Stefano Teso, Luc De Raedt. EURO 2019. cl

  • Toward Faithful Explanatory Active Learning with Self-explainable Neural Nets. Stefano Teso. Workshop on Interactive Adaptive Learning @ ECML-PKDD 2019. [pdf, code] iml xai

2018

  • Constructive Preference Elicitation. Paolo Dragone, Stefano Teso, Andrea Passerini Frontiers in Robotics and AI. [pdf] pe cl

  • Learning Constraints from Examples. Luc De Raedt, Andrea Passerini, Stefano Teso. AAAI 2018. [pdf] cl

  • Constructive preference elicitation over hybrid combinatorial spaces. Paolo Dragone, Stefano Teso, Andrea Passerini. AAAI 2018. [pdf] pe cl

  • Decomposition Strategies for Constructive Preference Elicitation. Paolo Dragone, Stefano Teso, Mohit Kumar, Andrea Passerini. AAAI 2018. [pdf] pe cl

  • Learning SMT(LRA) constraints using SMT solvers. Samuel Kolb, Stefano Teso, Andrea Passerini, Luc De Raedt. IJCAI 2018. [pdf] cl

  • Elements of an Automatic Data Scientist. Luc De Raedt, Hendrick Blockeel, Samuel Kolb, Stefano Teso, Gust Verbruggen. IDA 2018. [paper] cl

  • Automating Layout Synthesis with Constructive Preference Elicitation. Luca Erculiani, Paolo Dragone, Stefano Teso, Andrea Passerini. ECML-PKDD 2018. [paper] pe cl

  • Pyconstruct: constraint programming meets structured prediction Paolo Dragone, Stefano Teso, Andrea Passerini. IJCAI 2018 (Demo Track). [pdf] srl pe cl

  • Constraint Learning using Tensors. Mohit Kumar, Stefano Teso, Patrick De Causmaecker, Luc De Raedt. EURO 2018. cl

2017

  • Structured Learning Modulo Theories. Stefano Teso, Roberto Sebastiani, Andrea Passerini. Artificial Intelligence. [paper] srl

  • Investigating the Association between Social Interactions and Personality States Dynamics D Gundogdu, AN Finnerty, J Staiano, Stefano Teso, Andrea Passerini, F Pianesi, … Royal Society Open Science. [paper] other

  • Coactive Critiquing: Elicitation of Preferences and Features. Stefano Teso, Paolo Dragone, Andrea Passerini. AAAI 2017. [pdf] pe cl

  • Constructive Preference Elicitation for Multiple Users with Setwise Max-margin. Stefano Teso, Andrea Passerini, Paolo Viappiani. Algorithmic Decision Theory 2017. [paper] pe cl

2016

  • Rule Mining in Feature Space. Stefano Teso, Andrea Passerini. [preprint] nesy cl

  • Constructive Preference Elicitation by Setwise Max-Margin Learning. Stefano Teso, Andrea Passerini, Paolo Viappiani. IJCAI 2016. [pdf] pe cl

  • Structured feedback for preference elicitation in complex domains. Stefano Teso, Paolo Dragone, Andrea Passerini. BeyondLabeler Workshop @ IJCAI 2016. [pdf] pe cl

  • Constructive layout synthesis via coactive learning Paolo Dragone, Luca Erculiani, Maria Teresa Chietera, Stefano Teso, Andrea Passerini. Constructive Machine Learning Workshop @ NeurIPS 2016. [pdf] pe cl

2015

  • Inducing Sparse Programs for Learning Modulo Theories. Stefano Teso, Andrea Passerini. Constructive Machine Learning Workshop @ ICML 2015. [pdf] srl

  • Constructive Learning Modulo Theories. Stefano Teso, Roberto Sebastiani, Andrea Passerini. Constructive Machine Learning Worshop @ ICML 2015. [pdf] srl

  • Predicting Virus Mutations through Statistical Relational Learning. Elisa Cilia, Stefano Teso, Sergio Ammendola, Tom Lenaerts, Andrea Passerini. BMC Bioinformatics. [pdf] srl

2014

  • Joint Probabilistic-Logical Refinement of Multiple Protein Feature Predictors. Stefano Teso, Andrea Passerini. BMC bioinformatics. [paper] srl

  • Improved Multi-level Protein–Protein Interaction Prediction with Semantic-Based Regularization. Claudo Sacca, Stefano Teso, Michelangeli Diligenti, Andrea Passerini BMC bioinformatics. [paper] srl

  • Hybrid Statistical Relational Learning with Optimization Modulo Theories. Stefano Teso, Roberto Sebastiani, Andrea Passerini. Constructive Machine Learning Workshop @ NeurIPS 2014. [pdf] srl

2013

  • Statistical relational learning for proteomics: function, interactions and evolution. Stefano Teso. Ph.D. Thesis. [pdf] srl

  • Ego-centric graphlets for personality and affective states recognition. Stefano Teso, Jacopo Staiano, Bruno Lepri, Andrea Passerini, Fabio Pianesi. International Conference on Social Computing 2013. [paper] other

2012

  • Predicting virus mutations through relational learning. Elisa Cilia, Stefano Teso, Sergio Ammendola, Tom Lenaerts, Andrea Passerini. AIMM 2012. [pdf] srl

2010

  • An on/off lattice approach to protein structure prediction from contact maps. Stefano Teso, Cristina Di Risio, Andrea Passerini, Roberto Battiti. IAPR International Conference on Pattern Recognition in Bioinformatics 2010. [pdf] other

  • From on-going to complete activity recognition exploiting related activities. Carlo Nicolini, Bruno Lepri, Stefano Teso, Andrea Passerini. International Workshop on Human Behavior Understanding 2010. [paper] other

2008

  • Notes on stochastic simulation of chemical kinetics with cycle-leaping. Stefano Teso, Paola Lecca. Technical Report. [pdf] other