I am a Postdoctoral Researcher at the Neuro-Symbolic AI Group at the Idiap Research Institute.
My main research activity lies at the intersection of Natural Language Processing (NLP), reasoning, and explanation, investigating the development of Artificial Intelligence (AI) models that can reason over natural language in complex domains (e.g., science, mathematics, clinical and biomedical applications, ethical reasoning) and generate explanations in support of their decisions.
Specifically, my research attempts to inject explicit knowledge into hybrid neuro-symbolic architectures to improve the robustness, controllability and faithfulness of explanations for neural models (including Large Language Models) and, in the long term, to understand the principles governing explanatory reasoning in humans.
Moreover, I am interested in developing evaluation methodologies to interpret and uncover inference mechanisms and limitations of deep learning models, with a focus on symbolic, logical reasoning and out-of-distribution generalisation.
Main Research Interests. Natural Language Processing, Natural Language Explanations, Explanatory Reasoning, Neuro-Symbolic Models and Representations, Interpretability and Causal Analysis of NLP Models.
News
- February 2025. I gave a talk titled "Reasoning with Natural Language Explanations: From Epistemology to Neuro-Symbolic AI" at the Explainable AI Seminars, Imperial College London.
- January 2025. Eliciting Critical Reasoning in Retrieval-Augmented Language Models via Contrastive Explanations has been accepted to NAACL 2025.
- January 2025. SylloBio-NLI: Evaluating Large Language Models on Biomedical Syllogistic Reasoning has been accepted to NAACL 2025.
- January 2025. Controlling Equational Reasoning in Large Language Models with Prompt Interventions has been accepted to AAAI 2025.