| Task Name | Table Type | Description (related work) | Example Dataset |
|---|---|---|---|
| Table QA | WT | Answer a NL question given a table ( [28, 29]) | WikiTableQuestion [30] |
| Tablefact verification | WT | Verifying facts given a table ( [29, 31]) | TabFact [32] |
| Table-to-text | WT | Produce a NL question given a table ( [11]) | ToTTo [33] |
| Data cleaning | WT/SS/DB | Correct errors of table data ( [34, 35]) | - |
| Column/Row/Cellpopulation | WT/SS/DB | Populate possible column/row/cell for a table ( [11, 12]) | TURL [9] |
| Entity linking | WT | Link the selected entity to the knowledge base ( [11, 12]) | TURL [9] |
| Column typeannotation | WT | Choose types for the column in the table ( [11, 12]) | TURL [9] |
| Spreadsheetmanipulation | SS | Manipulate spreadsheets ( [16, 36]) | SpreadsheetBench [37] |
| NL2SQL | DB | Translate a NL question to a SQL query ( [38, 39]) | Spider [40] |
| Data analysis | SS/DB | Table data analysis pipeline, consists offeature engineering, machine learning, etc. ( [41, 42]) | DS-1000 [43] |
| Table detection | DOC | Locate tables in documents ( [44]) | TableBank [45] |
| Table extraction | DOC | Extract and structuralize tables from documents ( [46, 44]) | PubTabNet [47] |
有価証券報告書の読み取り¶
UFO 2024（有報読み取りコンペ）¶
背景： 有価証券報告書の表を対象としたコンペティションの提案
タスク：
Table Retrievalタスク：質問に該当する表を検索する
Table QAタスク：単一の表から、質問に該当するセルのID or 値を取り出すタスク