ResearchLoom (Samhita) Documentation & Wiki
Architecture, Stylus Dynamics, Flow Maps, and Systematic Literature Synthesis
Project maintained by sohamsans
Hosted on GitHub Pages — Theme by mattgraham
Autonomous Flow Maps & Knowledge Graphs
Flow Maps provide an interactive visual workspace to explore research papers, group findings, link notes, and generate autonomous knowledge graphs.
🤖 AI-Generated Flow Maps
ResearchLoom can autonomously construct a comprehensive knowledge graph directly from your research queries or active paper candidate pools:
- In Flow Maps, click the AI Map button on the top canvas toolbar.
- Enter your research domain or topic (e.g., “Self-Supervised Vision Transformers for Histopathology”).
- The backend evaluates key concepts, methodologies, bottlenecks, and benchmarks using your configured LLM (Gemini, Claude, Groq, or OpenAI).
- An interconnected graph of Core Problems, Methodological Paradigms, Key Citations, and Open Research Gaps is generated and placed onto the canvas with automatic force-directed spacing.
👆 Tablet Multi-Touch & Gesture Navigation
- One-Finger Drag: Select and move paper cards, folder nodes, or sticky notes across the canvas.
- Two-Finger Pan: Drag anywhere on the background to pan the canvas viewport smoothly.
- Two-Finger Pinch-to-Zoom: Fluid zoom in/out with hardware acceleration.
- Export to Stylus Studio: Click the arrow icon on any paper card to export its abstract or key formula directly to Stylus Studio for mathematical derivation and note-taking.