LOOM: Personalized Learning Informed by Daily LLM Conversations Toward Long-Term Mastery via a Dynamic Learner Memory Graph

TL;DR

LOOM addresses the need for learning systems that balance continuity and initiative by deriving evolving learner needs from daily LLM conversations and organizing them into a dynamic memory graph. It introduces a four-stage agentic pipeline—conversation observation and summarization, topic decision and outline proposal, full course content generation, and progress tracking with memory updates—to transform everyday chats into personalized learning trajectories. A formative study with ten participants suggests LOOM yields relevant, coherent lessons and helps surface knowledge gaps, though improvements are needed for content depth and user control. These findings point to design implications for robust, mixed-initiative learner pipelines that integrate structured learner modeling with everyday interactions, enabling sustained long-term mastery.

Abstract

Foundation models are increasingly used to personalize learning, yet many systems still assume fixed curricula or coarse progress signals, limiting alignment with learners' day-to-day needs. At the other extreme, lightweight incidental systems offer flexible, in-the-moment content but rarely guide learners toward mastery. Prior work privileges either continuity (maintaining a plan across sessions) or initiative (reacting to the moment), not both, leaving learners to navigate the trade-off between recency and trajectory-immediate relevance versus cumulative, goal-aligned progress. We present LOOM, an agentic pipeline that infers evolving learner needs from recent LLM conversations and a dynamic learner memory graph, then assembles coherent learning materials personalized to the learner's current needs, priorities, and understanding. These materials link adjacent concepts and surface gaps as tightly scoped modules that cumulatively advance broader goals, providing guidance and sustained progress while remaining responsive to new interests. We describe LOOM's end-to-end architecture and working prototype, including conversation summarization, topic planning, course generation, and graph-based progress tracking. In a formative study with ten participants, users reported that LOOM's generated lessons felt relevant to their recent activities and helped them recognize knowledge gaps, though they also highlighted needs for greater consistency and control. We conclude with design implications for more robust, mixed-initiative learning pipelines that integrate structured learner modelling with everyday LLM interactions.

Figures (2)

• Figure 1: The LOOM pipeline unifies initiative and continuity for personalized learning: Step 1: Conversation Observation & Summarization surfaces recurring themes and gaps in user's chats with an LLM assistant; A learner graph organizes mentioned concepts and demonstrated prior knowledge and learning progress; Step 2: Topic Decision & Outline Generation identifies learning topic from user inquiries and potential adjacent concept to explore, displayed as course outline proposed to the user; Step 3: Course Content Generation serves adaptive modules linked to adjacent concepts and current goals; Step 4: Updated & Regroup Learner Graph records user engagement and learning outcomes to update mastery progress and guide next steps and reinforcement over time.
• Figure 2: Evaluation responses: stacked distributions of participants' Likert responses (colors indicate response bands from Strongly Disagree to Strongly Agree). Q2 is reverse-coded, so higher scores indicate fewer repetitions (a more positive outcome).