LLMs as Stratification Signals for KG Accuracy Evaluation
Summary: Uses aggregated LLM predictions as stratification signals—not truth estimators—for statistically guaranteed, lower-cost KG accuracy sampling. Distillation transfers these signals to student models, cutting annotation to 0.25% of facts and reducing costs 11–54% across six large KGs. (summarized by gpt-5.6-luna on Aug 17 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Stefano Marchesin (University of Padua)
- 2. Matteo Ceccarello (University of Padua)
- 3. Gianmaria Silvello (University of Padua)
BibTeX Citation
@article{marchesin_vldb26,
title = {{LLMs as Stratification Signals for KG Accuracy Evaluation}},
author = {Marchesin, Stefano and Ceccarello, Matteo and Silvello, Gianmaria},
journal = {PVLDB},
series = {{VLDB} '26},
volume = {19},
number = {9},
pages = {2005--2018},
doi = {10.14778/3819518.3819530},
url = {https://doi.org/10.14778/3819518.3819530},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 7 of 7 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,373 | High-Throughput Vector Similarity Search in Knowledge Graphs | 2023 | SIGMOD | 0.00010891169 |
| 2,429 | Building, Maintaining, and Using Knowledge Bases: A Report from the Trenches | 2013 | SIGMOD | 8.4775762e-05 |
| 3,015 | Saga: A Platform for Continuous Construction and Serving of Knowledge At Scale | 2022 | SIGMOD | 7.7504975e-05 |
| 4,904 | Growing and Serving Large Open-domain Knowledge Graphs | 2023 | SIGMOD | 6.3605903e-05 |
| 5,385 | Efficient Knowledge Graph Accuracy Evaluation | 2019 | VLDB | 6.1529688e-05 |
| 9,490 | Credible Intervals for Knowledge Graph Accuracy Estimation | 2025 | SIGMOD | 5.168414e-05 |
| 11,578 | Efficient and Reliable Estimation of Knowledge Graph Accuracy | 2024 | VLDB | 4.9769913e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 9,251 | Deep Active Alignment of Knowledge Graph Entities and Schemata | 2023 | SIGMOD |
| 2 | 11,170 | ScaleLLM: A Technique for Scalable LLM-augmented Data Systems | 2025 | SIGMOD |
| 3 | 10,538 | In-context Clustering-based Entity Resolution with Large Language Models: A Design Space Exploration | 2026 | SIGMOD |
| 4 | 9,490 | Credible Intervals for Knowledge Graph Accuracy Estimation | 2025 | SIGMOD |
| 5 | 3,467 | In-depth Analysis of Graph-based RAG in a Unified Framework | 2025 | VLDB |
| 6 | 10,583 | GeoKGM: A Multimodal Large Language Model for Zero-Shot Knowledge Graph Completion in Geospatial Databases | 2026 | SIGMOD |
| 7 | 11,578 | Efficient and Reliable Estimation of Knowledge Graph Accuracy | 2024 | VLDB |
| 8 | 9,486 | Dialogue Benchmark Generation from Knowledge Graphs with Cost-Effective Retrieval-Augmented LLMs | 2025 | SIGMOD |
| 9 | 8,526 | Cut Costs, Not Accuracy: LLM-Powered Data Processing with Guarantees | 2026 | SIGMOD |
| 10 | 5,385 | Efficient Knowledge Graph Accuracy Evaluation | 2019 | VLDB |