PersonaPath benchmark for Knowledge-Centric learning path planning

Product or Programme · All products and programmes
Evidence · 6 claims (6 for) · 1 study (1 design), q2 · 0 of 1 report an effect size · 6 claims rest on one study

Description

A benchmark introduced by Yu Liu et al. that pairs learner personas and mastery states with a textbook knowledge graph to evaluate prerequisite-aware, adaptive learning-path generation.

Components

Claims

• Strongest LLMs reach only about 29.5% final pass rate on Basic Education KC planning and drop to 14.6% on Higher Education [+W]
• One-shot path generation raises Validity but lowers Adaptivity and Efficiency relative to step-by-step interaction [+W]
• Adaptivity to individual learners is the primary bottleneck for LLMs in KC path planning [+W]
• Removing explicit mastery information from learner profiles mainly degrades Adaptivity [+W]
• Contextual noise from irrelevant textbooks broadly degrades KC planning, most in Validity and Efficiency [+W]
• Retrieval augmentation raises path Validity but not the joint Final Pass rate [+W]

Related Products and Programmes

• Teaching Monster Challenge benchmark
• The Pedagogy Benchmark

Key Sources

• Yu Liu, Zeming Liu, Tianle Zhang, Zihao Cheng, Yuhang Guo, Kehai Chen, Min Zhang, Yunhong Wang, Haifeng Wang. (2026). PersonaPath: Towards Knowledge-Centric Personalized Learning Path Planning. arXiv. https://arxiv.org/abs/2609.18861

Page metadata

Field	Value
type	product
id	personapath-benchmark-for-knowledge-centric-learning-path-planning
title	PersonaPath benchmark for Knowledge-Centric learning path planning
description	A benchmark introduced by Yu Liu et al.
status	review
product_kind	dataset
generated	by process:settle-candidates on 2026-10-11
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.