Primary supervisor
Zhixi CaiCo-supervisors
Foundation models have shown strong reasoning abilities, but their reasoning process is often implicit, difficult to inspect, and not always reliable. This project explores neuro-symbolic reasoning as a way to make AI reasoning more structured, interpretable, and robust.
The project will study how neural models can be combined with explicit logic-based reasoning. A neural model may be used to understand or decompose a problem, while a symbolic reasoning component performs structured inference using logic, rules, programs, or other formal representations.
Aim/outline
The aim of this project is to investigate how explicit logic can improve the reliability and interpretability of AI reasoning systems.
The student will review related work, select a suitable reasoning task, implement a neuro-symbolic reasoning prototype, and compare it with a neural-only baseline. The final project should include an experimental evaluation and analysis of the strengths and limitations of the proposed approach.
Required knowledge
Good Python programming skills, basic machine learning knowledge, and interest in AI reasoning.
Experience with large language models, logic, symbolic AI, program execution, or evaluation of AI systems.