Methods-entity knowledge graph improves academic survey generation
SurveyGraph: Generating Academic Surveys via Methods-Entity Knowledge Graphs
Computation and LanguageArtificial IntelligenceInformation Retrieval
Summary
Creating summaries of academic research is hard because papers are usually treated as simple documents without showing how ideas connect. The authors created SurveyGraph, which uses a special graph to link research methods and entities, showing how they build on or compete with each other. This graph helps find key papers, organize content better, and create more accurate and structured summaries. Tests show this method is better at finding relevant papers and generating supported sentences than traditional approaches.
What this means in practice
- •For academic content platform developers: Build tools that generate structured, evidence-backed academic literature surveys by linking research methods and entities through a knowledge graph.
- •For enterprise knowledge management teams: Implement graph-based retrieval systems to improve discovery and organization of technical documents with complex method-entity relationships.
Authors
Yifan Yuan, Yancheng Zhu, Zuoyu Zhang, Xiaogang Peng
Abstract
Most automated survey generation systems retrieve papers as flat documents, limiting their ability to capture research structure. We present SurveyGraph, a framework built around a Methods-Entity Knowledge Graph (MEKG) encoding method-level follow-on, competition, and shared-resource relations. The graph supports core-paper discovery, retrieval, hybrid outlining, and evidence-gated narrative synthesis. We evaluate retrieval, semantic edges, citation grounding, collection coverage, and generation-stage components. In a controlled seven-method equal-budget benchmark, MEKG-anchored fusion balances recall and structural density, with an independent co-citation check. In a separate archived graph-native diagnostic, graph-aware retrieval nearly doubles recall over flat lexical baselines across eight runs (p = 0.004). A semantic evidence gate achieves AUC = 0.75. Audits of 160 generated sentences find 80.6% anchored-or-partial citation support. Human adjudication of 108 stratified semantic edges finds 68.5% supported-or-partial relations (binary-support Cohen's kappa = 0.64), with improves strongest (80.6%) and extends weakest (52.8%). A SurveyBench-facing collection probe reaches mean reference coverage of 0.521 across four topics with an eight-paper budget. Same-backend Qwen diagnostics on a 30-paper/74-relation graph show reduced organization signals when hybrid outlining is removed. Concurrent-system comparisons provide bounded context rather than matched-backend end-to-end evidence. The results position MEKG as an auditable structural layer for typed diagnostics, relation-specific filtering, and controlled survey generation. Code and artifacts: https://github.com/xzzvsxd/SurveyGraph.