{"url":"/dataset/wikitables-turl","name":"WikiTables-TURL","full_name":null,"description_markdown":"The WikiTables-TURL dataset was constructed by the authors of [TURL](https://paperswithcode.com/paper/turl-table-understanding-through) and is based on the WikiTable corpus, which is a large collection of Wikipedia tables. The dataset consists of 580,171 tables divided into fixed training, validation and testing splits. Additionally, the dataset contains metadata about each table, such as the table name, table caption  and column headers. \r\n\r\n406,706 of these tables are annotated for the Column Type Annotation (CTA) task, 55,970 tables for the Columns Property Annotation (CPA) task and 200,744 tables for the Cell Entity Annotation (CEA) task. As classes for the CTA and CPA, Freebase's types and relations were used, whereas for the CEA task entities from Freebase were used. The table below lists the total annotated columns (or cells in the case of CEA) for each split and for each task  as well as the number of classes used for annotation.\r\n\r\n|     | Training | Validation | Testing | Classes |\r\n|-----|--------|----------|-------|-------|\r\n| CTA | 628,254 |13,391| 13,025 | 255 |\r\n| CPA | 62,954| 2,175 | 2,072 | 121 |\r\n| CEA | 1,264,217 | 76,720  | 225,777 | 1,787,737 |\r\n\r\nThe authors have made the dataset and its variants publicly available for [download](https://buckeyemailosu-my.sharepoint.com/personal/deng_595_buckeyemail_osu_edu/_layouts/15/onedrive.aspx?id=%2Fpersonal%2Fdeng%5F595%5Fbuckeyemail%5Fosu%5Fedu%2FDocuments%2FBuckeyeBox%20Data%2FTURL&ga=1).","description_withheld":null,"homepage":"","introduced_date":"2020-06-26","introduced_date_note":null,"introduced_by":{"paper":"/paper/turl-table-understanding-through","title":"TURL: Table Understanding through Representation Learning","first_author":"Xiang Deng","url":null},"license":null,"modalities":[{"name":"Tabular","url":"/datasets/modality/tabular"}],"tasks":[{"name":"Data Integration","url":"/task/data-integration","datasets_with_task":"/datasets/task/data-integration"},{"name":"Table annotation","url":"/task/table-annotation","datasets_with_task":"/datasets/task/table-annotation"},{"name":"Column Type Annotation","url":"/task/column-type-annotation","datasets_with_task":"/datasets/task/column-type-annotation"},{"name":"Cell Entity Annotation","url":"/task/cell-entity-annotation","datasets_with_task":"/datasets/task/cell-entity-annotation"},{"name":"Columns Property Annotation","url":"/task/columns-property-annotation","datasets_with_task":"/datasets/task/columns-property-annotation"}],"languages":[],"variants":["WikiTables-TURL","WikiTables-TURL-CTA","WikiTables-TURL-CPA","WikiTables-TURL-CEA"],"data_loaders":[],"num_papers_in_archive":7,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/column-type-annotation-on-wikitables-turl-cta","task":"Column Type Annotation","dataset_variant":"WikiTables-TURL-CTA","rows":3,"metrics":["F1 (%)","Macro-F1"],"first_row_in_archive_order":{"model":"TURL","paper":"/paper/turl-table-understanding-through","metrics":{"F1 (%)":"94.75"},"code_links":[{"title":"sunlab-osu/TURL","url":"https://github.com/sunlab-osu/TURL"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/columns-property-annotation-on-wikitables","task":"Columns Property Annotation","dataset_variant":"WikiTables-TURL-CPA","rows":3,"metrics":["F1 (%)"," Macro-F1"],"first_row_in_archive_order":{"model":"TURL","paper":"/paper/turl-table-understanding-through","metrics":{"F1 (%)":"94.91"},"code_links":[{"title":"sunlab-osu/TURL","url":"https://github.com/sunlab-osu/TURL"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/cell-entity-annotation-on-wikitables-turl-cea","task":"Cell Entity Annotation","dataset_variant":"WikiTables-TURL-CEA","rows":1,"metrics":["F1 (%)"],"first_row_in_archive_order":{"model":"TURL","paper":"/paper/turl-table-understanding-through","metrics":{"F1 (%)":"68"},"code_links":[{"title":"sunlab-osu/TURL","url":"https://github.com/sunlab-osu/TURL"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/watchog-a-light-weight-contrastive-learning","title":"Watchog: A Light-weight Contrastive Learning based Framework for Column Annotation","date":"2023-12-12","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/annotating-columns-with-pre-trained-language","title":"Annotating Columns with Pre-trained Language Models","date":"2021-04-05","rows_on_this_dataset":2,"code_links":1,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":3,"samples_ran":3,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/turl-table-understanding-through","title":"TURL: Table Understanding through Representation Learning","date":"2020-06-26","rows_on_this_dataset":3,"code_links":1,"syntology":{"read_at":"2026-09-25T09:33:49+00:00","samples_harvested":1,"samples_ran":1,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-25T09:33:49+00:00","papers_with_samples":2,"samples_harvested":4,"samples_ran":4,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}