Search and Filter

Proactive Agentic Whiteboards: Enhancing Diagrammatic Learning

Date

AI Overview

Why it matters

Creating clear diagrams in real time while speaking during lectures is cognitively demanding, and incomplete visual aids may hinder student comprehension.

Study at a glance
  • Question: Can an AI whiteboard assistant proactively complete and refine educational diagrams based on spoken explanations and partial sketches?
  • Setting and data: Four simulated teaching sessions recorded with one practitioner across computer science, web development, biology, and chemistry topics.
  • Method: System development and demonstration across four simulated educational scenarios.
  • AI tools: DrawDash, gemini-2.5-flash-image
Key findings
  • DrawDash successfully turned rough sketches into structured diagrams across computer science, biology, and chemistry scenarios using speech and visual cues.
  • The system generated erroneous connections in a multi-step web development flow and simplified fine-grained details in a chemistry diagram.
  • Layout issues occurred during generation, including overlapping labels and arrows in a biology diagram, highlighting challenges in fine spatial adjustment.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

Educators frequently rely on diagrams to explain complex concepts during lectures, yet creating clear and complete visual representations in real time while simultaneously speaking can be cognitively demanding. Incomplete or unclear diagrams may hinder student comprehension, as learners must mentally reconstruct missing information while following the verbal explanation. Inspired by advances in code completion tools, we introduce DrawDash, an AI-powered whiteboard assistant that proactively completes and refines educational diagrams through multimodal understanding. DrawDash adopts a TAB-completion interaction model: it listens to spoken explanations, detects intent, and dynamically suggests refinements that can be accepted with a single keystroke. We demonstrate DrawDash across four diverse teaching scenarios, spanning topics from computer science and web development to biology. This work represents an early exploration into reducing instructors' cognitive load and improving diagram-based pedagogy through real-time, speech-driven visual assistance, and concludes with a discussion of current limitations and directions for formal classroom evaluation.