PhD defense Zacchary Sadeddine: Meaning Representation Frameworks and Reasoning in the Era of LLMs
Télécom Paris, 19 place Marguerite Perey F-91120 Palaiseau [getting there], amphi Estaunié and in videoconferencing
Jury
- Maxime AMBLARD, Professor, Université de Lorraine, France (Reviewer – Jury President)
- Anthony HUNTER, Professor, University College London, United Kingdom (Reviewer)
- Serena VILLATA, Senior Researcher, CNRS, France (Examiner)
- Emily ALLAWAY, Chancellor’s Fellow, University of Edinburgh, United Kingdom (Examiner)
- Kyle RICHARDSON, Senior Research Scientist, Allen Institute for Artificial Intelligence, USA (Examiner)
- Roberto NAVIGLI, Professor, Università Sapienza di Roma, Italy (Examiner)
- Fabian SUCHANEK, Professor, Télécom Paris, France (PhD Supervisor)
Abstract
Large Language Models (LLMs) are now used for a wide range of tasks, many of which require reasoning abilities. However, these abilies remain limited and lack transparency. This thesis explores how to improve the reasoning, transparency and robustness of LLMs by integrating symbolic structures.
First, we conduct an analysis of the societal issues arising from the new role of LLMs in our access to knowledge. Fifteen major issues are identified, as well as current and potential migation strategies, drawing on both technical solutions and regulatory approaches.
… which encode the semantics of natural language into graph structures. A comprehensive survey of MRFs is presented, introducing a new classification based on their structural properties, as well as the available resources, empirical use and research directions. This repositions MRFs as computational artifacts capable of complementing neural models in complex tasks. Building upon this, the thesis introduces VANESSA, a neuro-symbolic reasoning system that integrates MRFs with LLMs, as well as a new representation and symbolic parsing process. VANESSA uses this representation to decompose reasoning problems into three simpler subtasks: parsing, natural language inference (NLI) and formal solving. Experimental results show that VANESSA achieves performance comparable to LLMs on logical reasoning tasks, while producing outputs that are traceable and explainable, illustrating the added value of hybrid architectures. Finally, the thesis addresses the problem of step-by-step verification of reasoning chains, which are produced by LLMs. A novel benchmark of nearly 5,000 annotated reasoning steps is presented, assessing both logical validity and factual correctness. Though LLMs are able to detect some errors, neuro-symbolic approaches such as VANESSA achieve comparable performance while providing valuable transparency. Overall, the thesis advocates a hybrid vision of language-based arficial intelligence, where LLMs and symbolic structures are not competing paradigms but complementary tools. It opens new perspectives towards AI systems that are not only powerful, but also responsible, trustworthy and interpretable, combining the flexibility of neural models with the rigor of symbolic reasoning.