MindTrellis: Co-Creating Knowledge Structures with AI through Interactive Visual Exploration
AI and users jointly build dynamic knowledge graphs from documents
Knowledge workers often struggle to synthesize information from multiple documents into a structured conceptual understanding. Current tools are limited: LLM-based systems let users query information but not shape how knowledge is organized, while manual tools like mind maps support structure creation but lack intelligent assistance. Researchers from multiple institutions developed MindTrellis to address this gap.
MindTrellis is an interactive visual system where users and AI collaboratively build a dynamic knowledge graph. Users can query the graph to retrieve document-grounded information, introduce new concepts, modify relationships, and reorganize the hierarchy. In a user study with 12 participants creating slide decks, MindTrellis outperformed retrieval-only baselines in knowledge organization and cognitive load, as measured by expert ratings of content coverage and structural quality. The system represents a significant step toward more effective human-AI collaboration in knowledge work.
- MindTrellis enables users and AI to co-create dynamic knowledge graphs from documents
- Users can query the graph, add concepts, modify relationships, and reorganize hierarchies
- In a study of 12 participants creating slide decks, it outperformed retrieval-only baselines in organization and cognitive load
Why It Matters
Enables deeper, more flexible knowledge synthesis by combining human intuition with AI's processing power.