Research domains
- Problem–Solution Graphs & Inventive Design
- Knowledge Graphs & Causal Reasoning
- Scientific NLP & Information Extraction
- AI-Assisted Invention & Innovation
Nathan Witkowicz works on using generative AI to turn text into structured models that support invention, and designs the graph-building algorithm that turns texts into problem graphs: the engine of TRIZBench. Publications cover the extraction of causal maps from text with GPT-based question answering and the extraction of problem graphs, together with improvement perspectives, for AI-assisted invention.