Retrieve-and-Verify: A Table Context Selection Framework for Accurate Column Annotations
Summary: Introduce a retrieve-and-verify framework that selects compact, informative column contexts (instead of serializing whole tables) via unsupervised retrieval balancing relevance and diversity and role-aware target-context encoding (REVEAL). REVEAL+ adds a learned verification classifier with top-down inference to efficiently refine context subsets (quadratic search), yielding consistent CTA/CPA accuracy gains over SOTA across six benchmarks. (summarized by gpt-5-mini on Feb 11 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Zhihao Ding (Hong Kong Polytechnic University)
- 2. Yongkang Sun (Hong Kong Polytechnic University)
- 3. Jieming Shi (Hong Kong Polytechnic University)
BibTeX Citation
@inproceedings{ding_sigmod26,
title = {{Retrieve-and-Verify: A Table Context Selection Framework for Accurate Column Annotations}},
author = {Ding, Zhihao and Sun, Yongkang and Shi, Jieming},
series = {{SIGMOD} '26},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3769823},
url = {https://dl.acm.org/doi/10.1145/3769823},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Outgoing Citations (Sorted by Pagerank)
Showing 19 of 19 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 7,482 | Table Extraction and Understanding for Scientific and Enterprise Applications | 2020 | VLDB |
| 2 | 9,067 | Making Table Understanding Work in Practice | 2022 | CIDR |
| 3 | 4,515 | ArcheType: A Novel Framework for Open-Source Column Type Annotation using Large Language Models | 2024 | VLDB |
| 4 | 8,860 | Watchog: A Light-weight Contrastive Learning based Framework for Column Annotation | 2023 | SIGMOD |
| 5 | 11,186 | Unstructured Data Fusion for Schema and Data Extraction | 2024 | SIGMOD |
| 6 | 1,055 | Answering Table Queries on the Web using Column Keywords | 2012 | VLDB |
| 7 | 397 | TURL: Table Understanding through Representation Learning | 2021 | VLDB |
| 8 | 1,923 | Annotating Columns with Pre-trained Language Models | 2022 | SIGMOD |
| 9 | 753 | Recovering Semantics of Tables on the Web | 2011 | VLDB |
| 10 | 8,458 | RECA: Related Tables Enhanced Column Semantic Type Annotation Framework | 2023 | VLDB |