Reproduce “TrueReason: An Exemplar Personalised Learning System Integrating Reasoning with Foundational Models” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Sahan Bulathwela, Daniel Van Niekerk, Jarrod Shipton, Maria Perez-Ortiz, Benjamin Rosman, John Shawe-Taylor

TrueReason combines a conversational language model with specialist micro-skills for learner modeling and teaching activities. Its prototype uses TrueLearn to represent knowledge and interests, while components recommend resources and generate questions. In simulation, an AKS-sample DDPG variant marginally outperformed two alternatives; TopicQG scored 0.727 on WikiSemRel versus 0.615 for baseline. These component results do not establish learning gains in a system.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.