Problem
Research questions and friction points this paper is trying to address.
educational assistants
large language models
student engagement
resource synthesis
open-book learning
BLADE: Better Language Answers through Dialogue and Explanations
🤖 AI Summary
This work addresses a critical limitation of existing large language model–based educational assistants, which often provide direct answers without fostering students’ active engagement with course materials. To remedy this, the authors propose BLADE, a retrieval-augmented generation (RAG)–based conversational tutoring system that guides learners through dialogues grounded in explicit citations from course resources, prioritizing evidence-based interaction over immediate solution provision. Empirical results show that students prefer using BLADE when given a choice; quiz performance is highest when BLADE is used alone, while concurrent access to both BLADE and original materials significantly impairs learning—a counterintuitive finding suggesting that multiple resource use may impose excessive cognitive load. This study demonstrates a successful adaptation of RAG to educational contexts and underscores the importance of guided, traceable dialogue in enhancing learning outcomes.
📝 Abstract
Large language model (LLM)-based educational assistants often provide direct answers that short-circuit learning by reducing exploration, self-explanation, and engagement with course materials. We present BLADE (Better Language Answers through Dialogue and Explanations), a grounded conversational assistant that guides learners to relevant instructional resources rather than supplying immediate solutions. BLADE uses a retrieval-augmented generation (RAG) framework over curated course content, dynamically surfacing pedagogically relevant excerpts in response to student queries. Instead of delivering final answers, BLADE prompts direct engagement with source materials to support conceptual understanding. We conduct an impact study in an undergraduate computer science course, with different course resource configurations and show that BLADE improves students'navigation of course resources and conceptual performance compared to simply providing the full inventory of course resources. These results demonstrate the potential of grounded conversational AI to reinforce active learning and evidence-based reasoning.
Innovation
Methods, ideas, or system contributions that make the work stand out.
retrieval-augmented generation
dialogue-based learning
citation-grounded assistance
educational AI
cognitive load
🔎 Similar Papers
No similar papers found.
