Dynamite: Real-Time Debriefing Slide Authoring through AI-Enhanced Multimodal Interaction

2025-07-27

AI Summary

In ethics education, instructors face cognitive overload and time pressure when generating real-time presentation slides that accurately reflect both dominant themes and minority viewpoints emerging from small-group discussions. To address this, we propose an AI-augmented multimodal real-time slide editing system. Our approach innovatively integrates speech recognition, sketch understanding, and semantic analysis to establish semantic data binding and semantic suggestion mechanisms. The system supports dual-input modalities—speech and hand-drawn sketches—and enables context-aware, intent-driven linkage between pedagogical goals and dynamic discussion content, facilitating intelligent slide optimization. A user study with 12 participants demonstrates that our system significantly improves content accuracy and quality over a text-only AI baseline (p < 0.01), while enhancing slide organization and refinement efficiency by 47%.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Humans and AI: Learning Human Values and Preferences
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems

Abstract

Facilitating class-wide debriefings after small-group discussions is a common strategy in ethics education. Instructor interviews revealed that effective debriefings should highlight frequently discussed themes and surface underrepresented viewpoints, making accurate representations of insight occurrence essential. Yet authoring presentations in real time is cognitively overwhelming due to the volume of data and tight time constraints. We present Dynamite, an AI-assisted system that enables semantic updates to instructor-authored slides during live classroom discussions. These updates are powered by semantic data binding, which links slide content to evolving discussion data, and semantic suggestions, which offer revision options aligned with pedagogical goals. In a within-subject in-lab study with 12 participants, Dynamite outperformed a text-based AI baseline in content accuracy and quality. Participants used voice and sketch input to quickly organize semantic blocks, then applied suggestions to accelerate refinement as data stabilized.

Problem

• Real-time slide authoring is overwhelming during live discussions
• Instructors need to highlight common themes and rare viewpoints
• AI assistance improves accuracy and quality of debriefing slides

Innovation

• AI-assisted real-time slide updates
• Semantic data binding for dynamic content
• Voice and sketch input for rapid organization