add citation

README.md

@@ -1,18 +1,18 @@
----
-license: mit
-task_categories:
-- tabular-classification
-language:
-- ru
-tags:
-- tables
-- column-type-annotation
-- sti
-- russian
-pretty_name: Rutabert dataset
-size_categories:
-- 1M<n<10M
----
+---
+license: mit
+task_categories:
+- tabular-classification
+language:
+- ru
+tags:
+- tables
+- column-type-annotation
+- sti
+- russian
+pretty_name: RWT-RuTaBERT
+size_categories:
+- 1M<n<10M
+---
 
 # RWT-RuTaBERT
 
@@ -20,7 +20,7 @@ Dataset based on [Russian Web Tables](https://arxiv.org/abs/2210.06353) (RWT), w
 
 Only relational tables were chosen from _RWT_ with headers matching selected **170** DBpedia semantic types.
 
-Dataset contains `1.441.349` columns, and has fixed train / test split. 
+Dataset contains `1 441 349` columns, and has fixed train / test split. 
 
 | Split | Columns   | Tables  | Avg. columns per table |
 |-------|-----------| ------- | ---------------------- |
@@ -159,4 +159,20 @@ Dataset contains `1.441.349` columns, and has fixed train / test split.
 | борец        | 5           |
 | музей        | 4           |
 | банк         | 4           |
-| камера       | 4           |
+| камера       | 4           |
+
+## Citation
+
+```
+@INPROCEEDINGS{10659709,
+  author={Tobola, Kirill V. and Dorodnykh, Nikita O.},
+  booktitle={2024 Ivannikov Memorial Workshop (IVMEM)}, 
+  title={Semantic Annotation of Russian-Language Tables Based on a Pre-Trained Language Model}, 
+  year={2024},
+  volume={},
+  number={},
+  pages={62-68},
+  keywords={Annotations;Semantics;Layout;Training data;Knowledge graphs;Encyclopedias;Linguistics;tabular data;Russian-language tables;pre-trained language model;semantic table interpretation;column annotation;knowledge graph},
+  doi={10.1109/IVMEM63006.2024.10659709}
+}
+```