Assistant Professor
PI of DUCK Lab
PI of DUCK Lab
Eleonora
Giunchiglia
Eleonora Giunchiglia is an Assistant Professor at Imperial College London in the Department of Electrical and Electronic Engineering and a member of Imperial-X (I-X). She completed her DPhil at the University of Oxford in 2022, followed by a postdoctoral position at TU Wien before joining Imperial in 2024. Her research lies at the intersection of machine learning and formal reasoning, with a focus on neuro-symbolic AI — developing methods that integrate logical constraints and background knowledge into neural networks to make them safer, more reliable, and trustworthy, with applications ranging from constrained generative models to theorem proving and verifiable LLM reasoning.
Papers
-
Constraint-Satisfying Reinforcement Learning via a Neuro-Symbolic Projection LayerJacob Brand, Francesco Belardinelli, Eleonora Giunchiglia · NeSy · 2026
-
Can Transformers Perform LTLf Strategic Reasoning? SometimesGiuseppe De Giacomo, Antonio Di Stasio, Eleonora Giunchiglia, Eleni Kougioumtzi · NeSy · 2026
-
Can I Have Your Order? Monte-Carlo Tree Search for Slot Filling Ordering in Diffusion Language ModelsJoshua Ong Jun Leang, Yu Zhao, Mihaela Cătălina Stoian, Wenda Li, Shay B. Cohen, Eleonora Giunchiglia · ICML · 2026
-
PiCSAR: Probabilistic Confidence Selection And Ranking for Reasoning ChainsJoshua Ong Jun Leang, Zheng Zhao, Aryo Pradipta Gema, Sohee Yang, Wai-Chung Kwan, Xuanli He, Wenda Li, Pasquale Minervini, Eleonora Giunchiglia, Shay B. Cohen · ACL (Findings) · 2026
-
A Survey on Deep Learning Approaches for Tabular Data Generation: Utility, Alignment, Fidelity, Privacy, Diversity, and BeyondMihaela Cătălina Stoian, Eleonora Giunchiglia, Thomas Lukasiewicz · TMLR · 2026
-
Right for the Right Reasons: Avoiding Reasoning Shortcuts via Prototypical Neurosymbolic AILuca Andolfi, Eleonora Giunchiglia · NeurIPS · 2025
-
A Posteriori Verification or a Priori Design? Navigating Requirements-Driven Deep LearningEleonora Giunchiglia · ECAI · 2025
-
Beyond the convexity assumption: Realistic tabular data generation under quantifier-free real linear constraintsMihaela Cătălina Stoian, Eleonora Giunchiglia · ICLR · 2025
-
CCN+: A neuro-symbolic framework for deep learning with requirementsEleonora Giunchiglia, Alex Tatomir, Mihaela Cătălina Stoian, Thomas Lukasiewicz · International Journal of Approximate Reasoning · 2024
-
PiShield: A PyTorch Package for Learning with RequirementsMihaela Cătălina Stoian, Alex Tatomir, Thomas Lukasiewicz, Eleonora Giunchiglia · IJCAI · 2024
-
ULLER: A Unified Language for Learning and ReasoningEmile van Krieken, Samy Badreddine, Robin Manhaeve, Eleonora Giunchiglia · NeSy (Spotlight) · 2024
-
Deep generative models as an adversarial attack strategy for tabular machine learningSalijona Dyrmishi, Mihaela Cătălina Stoian, Eleonora Giunchiglia, Maxime Cordy · ICMLC · 2024
-
How Realistic Is Your Synthetic Data? Constraining Deep Generative Models for Tabular DataMihaela Cătălina Stoian, Salijona Dyrmishi, Maxime Cordy, Thomas Lukasiewicz, Eleonora Giunchiglia · ICLR · 2024
News
- 09-2026 Invited talk at NESYRE 2026 in Berlin Eleonora will give an invited talk at NESYRE 2026 — Neuro-symbolic AI, Mathematical Reasoning and Agents — at the Weierstrass Institute in Berlin (14–17 September). The speaker line-up looks amazing! ✨
- 08-2026 Tutorial at UAI 2026 in Amsterdam Mihaela and Eleonora will give a tutorial at UAI 2026 in Amsterdam (17–21 August): Beyond Soft Penalties: Hard Constraints for Neural Networks with Continuous Outputs — how to enforce hard constraints on neural networks with continuous outputs.
- 08-2026 Invited talk at TPM 2026 in Amsterdam Eleonora will give an invited talk at TPM 2026 — the 9th Workshop on Tractable Probabilistic Modeling — in Amsterdam on 21 August, co-located with UAI 2026. Looking forward to meeting this amazing community! ✨
- 07-2026 Two papers accepted at NeSy 2026! 🎉 We had two papers accepted at NeSy 2026, the International Conference on Neural-Symbolic Learning and Reasoning! Huge congratulations to Eleni and Jacob, who led these projects as MSc students. Jacob completed his MSc at Imperial, co-supervised by Eleonora and Francesco Belardinelli, while Eleni completed her MSc at Oxford, co-supervised by Eleonora, Giuseppe De Giacomo, and Antonio Di Stasio. ✨
- 07-2026 Eleonora on the future of AI panel at Imperial Eleonora took part in a panel on The future of AI: what’s next and what matters?, organised by the Imperial Policy Forum and chaired by Prof Tom Coates. The discussions were amazing!
- 06-2026 Invited talk at CILC 2026 Eleonora gave an invited talk at CILC 2026, the 41st Italian Conference on Computational Logic.
- 06-2026 Pythagoras-Prover released! 🚀 We are thrilled to announce the release of Pythagoras-Prover — a compute-efficient family of Lean theorem provers! Strong results at just 4B parameters, powered by Augmented Lean Formalisation. The models are now available on Hugging Face 🤗
- 05-2026 Talk at OxBridge Women in CS Conference On 9 May, Eleonora spoke at the OxBridge Women in CS Conference — a particularly special one for her, as she was herself an OxWoCS member during her PhD!
- 04-2026 Paper accepted at ICML 2026 Our paper Can I Have Your Order? Monte-Carlo Tree Search for Slot Filling Ordering in Diffusion Language Models has been accepted at ICML 2026.
- 03-2026 Paper accepted at ACL (Findings) Our paper PiCSAR: Probabilistic Confidence Selection And Ranking for Reasoning Chains has been accepted at ACL (Findings) 2026.
- 01-2026 Survey published in TMLR Our survey A Survey on Deep Learning Approaches for Tabular Data Generation has been published in Transactions on Machine Learning Research (TMLR).
- 10-2025 Frontiers in AI invited talk at ECAI 2025 Eleonora gave one of the Frontiers in AI invited talks at ECAI 2025.
- 09-2025 Paper accepted at NeurIPS 2025 Our paper Right for the Right Reasons: Avoiding Reasoning Shortcuts via Prototypical Neurosymbolic AI has been accepted at NeurIPS 2025.
- 08-2025 AI for Math Fund Eleonora has been awarded an AI for Math grant. Read more here
- 07-2025 Paper published at ECAI 2025 Eleonora’s paper A Posteriori Verification or a Priori Design? Navigating Requirements-Driven Deep Learning has been published at ECAI 2025.