Learn Before
Language-Matched Seeding as a Prerequisite for Graph-Expansion Gains
Scope Part 1: Strict-Parity Evaluation Contract for Graph-Aware (Auditable Strict-Parity Graph-RAG Paper)
Scope Part 1: Question-Disjoint and Target-Concept-Disjoint (Auditable Strict-Parity Graph-RAG Paper)
Scope Notes for Bidirectional Prerequisite Diffusion with Role-Aware (Auditable Strict-Parity Graph-RAG Paper)
Method-Related Scope Notes for Adaptive Depth Gating for Prerequisite Retrieval (Auditable Strict-Parity Graph-RAG Paper)
Three Durable Signals of Strict-Parity Prerequisite Retrieval
Under this paper's strict-parity contract, the corrective takeaway of the headline retrieval table is that adaptive depth gating is not the main story. The durable signals that survive strict-parity comparison are exactly three: (i) deterministic bidirectional diffusion with role-aware quotas on LectureBank-Full, (ii) language-matched seeding on MOOC-CS, and (iii) leakage-aware evaluation (question-disjoint and target-concept-disjoint controls) that narrows what these benchmarks can support. Adaptive depth and other graph-policy variants either remain statistically tied to a fixed-depth hierarchical baseline or do not contribute beyond these three signals.
0
1
Tags
Science
Auditable Strict-Parity Evaluation of Prerequisite-Graph Retrieval for RAG under Leakage Controls
Related
Three Durable Signals of Strict-Parity Prerequisite Retrieval
Methodological Narrowing of Graph-RAG Superiority Claims
Token-Cap Diagnostics Separated from Headline Comparisons
Reranker and Contract-Violation Reporting Policy in Strict-Parity Headlines
Three Durable Signals of Strict-Parity Prerequisite Retrieval
Evaluation-First Framing of Graph-Aware RAG Comparisons (Adopted from Zeng 2025, Han 2025, Ju 2025)
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
Three Durable Signals of Strict-Parity Prerequisite Retrieval
Diffusion Retriever Framed as a Controlled Test Case Inside the Evaluation Framework
LectureBank-Full Tight-Budget Advantage of Adaptive Depth Gating (Mean ΔR@k = +2.13 over k∈{1,2,3,4})
When Adaptive Depth Helps vs When Fixed-Depth Is Safer (Budget/Cap Guidance)
Three Durable Signals of Strict-Parity Prerequisite Retrieval
Adaptive Depth Gating Helps Only in Tight-k Settings on Headline Prerequisite Benchmarks
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)