Analysis-Section Scope Statement: Evidence Specific to Curated, Template-Based Prerequisite QA
The Analysis section makes an explicit scope statement about what the strict-parity evidence supports. Because the prerequisite datasets instantiate only two question families and the target-concept-disjoint test splits expose only unique held-out targets on LectureBank-Full and on MOOC-CS, paired-bootstrap intervals over per-target stability are necessarily wide and the statistical power to distinguish adaptive from hierarchical depth at is limited. The paper therefore treats the evidence as specific to curated, template-based prerequisite QA, rather than as a claim about broad educational or open-domain graph retrieval. This scoping is the operational consequence of the limited held-out budget, and it bounds the generality of every headline retrieval claim in the paper.
0
1
Tags
Science
Auditable Strict-Parity Evaluation of Prerequisite-Graph Retrieval for RAG under Leakage Controls
Related
Generation as Context-Quality Diagnostic, Not a Headline Claim
Claims Explicitly Avoided: Auto-Judge Validation, Semantic-Evasion Effects, End-to-End QA Gains
Qualitative Manual Bundles Remain Non-Evidentiary Until Multi-Annotator Replication
Analysis-Section Scope Statement: Evidence Specific to Curated, Template-Based Prerequisite QA
HotpotQA External-Validity Probe: Adaptive Depth Does Not Transfer to a Denser Non-Prerequisite Graph (FullWiki-1k: Flat 93.4 / Hier 92.9 / Adaptive 94.0 R@10)
Small Held-Out Target Sets in Target-Disjoint Prerequisite Splits
Bounded Empirical Findings with Explicit Boundary Checks
Three Durable Signals of Strict-Parity Prerequisite Retrieval
Tied-at-R@10 as Non-Rejection of Equivalence Under a Power Bound, Not a Positive Equivalence Claim
Analysis-Section Scope Statement: Evidence Specific to Curated, Template-Based Prerequisite QA