Learn Before
Method Part 3: Strict-Parity Evaluation Contract for Graph-Aware (Auditable Strict-Parity Graph-RAG Paper)
Method Part 2: Multilingual Encoder + CJK Query Rewrite as a MOOC-CS Control (Auditable Strict-Parity Graph-RAG Paper)
Method Part 1: MOOC-CS Prerequisite Benchmark (Auditable Strict-Parity Graph-RAG Paper)
MOOC-CS Graph Gain Requires Language-Matched Controls
On the MOOC-CS prerequisite benchmark, graph-aware retrieval gains under the default English-only setup are limited, and become substantial only after introducing language-matched query and encoder controls (matching the language of the query and the encoder to the underlying course materials). The result is reported under the paper's strict-parity contract, which fixes the candidate pool, cutoff , matching rule, and split policy and varies only the language-matching condition, so the change in graph gain is attributable to language matching rather than to graph policy. The finding illustrates that graph-specific gains can be masked or inflated by interface choices that are unrelated to graph structure.
0
1
Tags
Science
Auditable Strict-Parity Evaluation of Prerequisite-Graph Retrieval for RAG under Leakage Controls
Related
LightRAG-Style Baseline in Strict-Parity Prerequisite Retrieval
KAG-Style Structural Baseline in Strict-Parity Prerequisite Retrieval
LectureBank-Full R@10 Gain from Diffusion and Role-Aware Quotas
MOOC-CS Graph Gain Requires Language-Matched Controls
SP+ Strict Parity Plus a Learned Reranker (Reported Separately)
MOOC-CS Graph Gain Requires Language-Matched Controls
MOOC-CS Graph Gain Requires Language-Matched Controls
MOOC-CS Configuration Used in Hierarchical Prerequisite RAG (225 Concepts, 516 Edges, 1,016 QA)
MOOC-CS Headline R@10 Numbers: LightRAG/Truncated-PPR 25.6-25.9 vs Adaptive/Hierarchical Tied at 23.1
MOOC-CS Target-Disjoint R@10 Result (n=114): Negative Case, Adaptive Tied
Template Stripping on MOOC-CS Raises Hierarchical R@10 from 23.1 to 26.5 (MiniLM Encoder)