Datasets:

Modalities:
Image
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
HugoLaurencon commited on
Commit
b617f72
1 Parent(s): 20ef85d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +697 -2
README.md CHANGED
@@ -926,6 +926,701 @@ configs:
926
  - split: train
927
  path: vsr/train-*
928
  ---
929
- # Dataset Card for "the_cauldron"
930
 
931
- [More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
926
  - split: train
927
  path: vsr/train-*
928
  ---
929
+ # Dataset Card for The Cauldron
930
 
931
+ ## Dataset description
932
+
933
+ The Cauldron is part of the Idefics2 release.
934
+
935
+ It is a massive collection of 50 vision-language datasets (training sets only) that were used for the fine-tuning of the vision-language model Idefics2.
936
+
937
+ ## Load the dataset
938
+
939
+ To load the dataset, install the library `datasets` with `pip install datasets`. Then,
940
+ ```
941
+ from datasets import load_dataset
942
+ ds = load_dataset("HuggingFaceM4/the_cauldron", "ai2d")
943
+ ```
944
+ to download and load the config `ai2d` for example.
945
+
946
+ ## Data fields
947
+
948
+ An example of a sample looks as follows:
949
+ ```
950
+ {
951
+ "images" = [PIL.Image]
952
+ "texts" = [
953
+ {
954
+ "user": "Question: How many actions are depicted in the diagram?\nChoices:\nA. 6.\nB. 4.\nC. 8.\nD. 7.\nAnswer with the letter.",
955
+ "assistant": "Answer: D",
956
+ "source": "TQA"
957
+ }
958
+ ]
959
+ }
960
+ ```
961
+
962
+ In `images`, there is a list of images, to be placed before the text.
963
+ In `texts`, there is a conversation between a user and an assistant about the images that is represented by a list of turns.
964
+
965
+ ## Stats about the datasets in The Cauldron
966
+
967
+ | Dataset | # images | # Q/A pairs | # tokens |
968
+ |----------------------|----------|-------------|------------|
969
+ | *General visual question answering* |
970
+ | VQAv2 | 82,772 | 443,757 | 1,595,929 |
971
+ | COCO-QA | 46,287 | 78,736 | 286,982 |
972
+ | Visual7W | 14,366 | 69,817 | 279,268 |
973
+ | A-OKVQA | 16,539 | 17,056 | 236,492 |
974
+ | TallyQA | 98,680 | 183,986 | 738,254 |
975
+ | OK-VQA | 8,998 | 9,009 | 38,853 |
976
+ | HatefulMemes | 8,500 | 8,500 | 25,500 |
977
+ | VQA-RAD | 313 | 1,793 | 8,418 |
978
+ | Captioning |
979
+ | LNarratives | 507,444 | 507,444 | 21,328,731 |
980
+ | Screen2Words | 15,730 | 15,743 | 143,103 |
981
+ | VSR | 2,157 | 3,354 | 10,062 |
982
+ | *OCR, document understanding, text transcription* |
983
+ | RenderedText | 999,000 | 999,000 | 27,207,774 |
984
+ | DocVQA | 10,189 | 39,463 | 337,829 |
985
+ | TextCaps | 21,953 | 21,953 | 389,658 |
986
+ | TextVQA | 21,953 | 34,602 | 181,918 |
987
+ | ST-VQA | 17,247 | 23,121 | 127,846 |
988
+ | OCR-VQA | 165,746 | 801,579 | 6,073,824 |
989
+ | VisualMRC | 3,027 | 11,988 | 168,828 |
990
+ | IAM | 5,663 | 5,663 | 144,216 |
991
+ | InfoVQA | 2,118 | 10,074 | 61,048 |
992
+ | Diagram image-to-text| 300 | 300 | 22,196 |
993
+ | *Chart/figure understanding* |
994
+ | Chart2Text | 26,985 | 30,242 | 2,852,827 |
995
+ | DVQA | 200,000 | 2,325,316 | 8,346,234 |
996
+ | VisText | 7,057 | 9,969 | 1,245,485 |
997
+ | ChartQA | 18,271 | 28,299 | 185,835 |
998
+ | PlotQA | 157,070 | 20,249,479 | 8478299.278|
999
+ | FigureQA | 100,000 | 1,327,368 | 3,982,104 |
1000
+ | MapQA | 37,417 | 483,416 | 6,470,485 |
1001
+ | *Table understanding* |
1002
+ | TabMWP | 22,729 | 23,059 | 1,948,166 |
1003
+ | TAT-QA | 2,199 | 13,215 | 283,776 |
1004
+ | HiTab | 2,500 | 7,782 | 351,299 |
1005
+ | MultiHiertt | 7,619 | 7,830 | 267,615 |
1006
+ | FinQA | 5,276 | 6,251 | 242,561 |
1007
+ | WikiSQL | 74,989 | 86,202 | 9,680,673 |
1008
+ | SQA | 8,514 | 34,141 | 1,894,824 |
1009
+ | WTQ | 38,246 | 44,096 | 6,677,013 |
1010
+ | *Reasoning, logic, maths* |
1011
+ | GeomVerse | 9,303 | 9,339 | 2,489,459 |
1012
+ | CLEVR-Math | 70,000 | 788,650 | 3,184,656 |
1013
+ | CLEVR | 70,000 | 699,989 | 2,396,781 |
1014
+ | IconQA | 27,315 | 29,859 | 112,969 |
1015
+ | RAVEN | 42,000 | 42,000 | 105,081 |
1016
+ | Inter-GPs | 1,451 | 2,101 | 8,404 |
1017
+ | *Textbook/academic questions* |
1018
+ | AI2D | 3,099 | 9,708 | 38,832 |
1019
+ | TQA | 1,496 | 6,501 | 26,004 |
1020
+ | ScienceQA | 4,985 | 6,218 | 24,872 |
1021
+ | *Differences between 2 images* |
1022
+ | NLVR2 | 50,426 | 86,373 | 259,119 |
1023
+ | GSD | 70,939 | 141,869 | 4,637,229 |
1024
+ | Spot the diff | 8,566 | 9,524 | 221,477 |
1025
+ | *Screenshot to code* |
1026
+ | WebSight | 500,000 | 500,000 | 276,743,299|
1027
+ | DaTikz | 47,974 | 48,296 | 59,556,252 |
1028
+
1029
+ ## Decontamination
1030
+
1031
+ The Cauldron contains only the train split of each sub-datasets.
1032
+ On top of that, we removed the few examples containing an image also present in the test splits of MMMU, MathVista or MMBench.
1033
+
1034
+ ## References to the original datasets
1035
+
1036
+ <details>
1037
+ <summary>References to the original datasets</summary>
1038
+
1039
+ @misc{AI2D,
1040
+ title={A Diagram Is Worth A Dozen Images},
1041
+ author={Aniruddha Kembhavi and Mike Salvato and Eric Kolve and Minjoon Seo and Hannaneh Hajishirzi and Ali Farhadi},
1042
+ year={2016},
1043
+ eprint={1603.07396},
1044
+ archivePrefix={arXiv},
1045
+ primaryClass={cs.CV}
1046
+ }
1047
+
1048
+ @misc{A-OKVQA,
1049
+ title={A-OKVQA: A Benchmark for Visual Question Answering using World Knowledge},
1050
+ author={Dustin Schwenk and Apoorv Khandelwal and Christopher Clark and Kenneth Marino and Roozbeh Mottaghi},
1051
+ year={2022},
1052
+ eprint={2206.01718},
1053
+ archivePrefix={arXiv},
1054
+ primaryClass={cs.CV}
1055
+ }
1056
+
1057
+ @inproceedings{Chart2Text,
1058
+ title = "Chart-to-Text: Generating Natural Language Descriptions for Charts by Adapting the Transformer Model",
1059
+ author = "Obeid, Jason and
1060
+ Hoque, Enamul",
1061
+ editor = "Davis, Brian and
1062
+ Graham, Yvette and
1063
+ Kelleher, John and
1064
+ Sripada, Yaji",
1065
+ booktitle = "Proceedings of the 13th International Conference on Natural Language Generation",
1066
+ month = dec,
1067
+ year = "2020",
1068
+ address = "Dublin, Ireland",
1069
+ publisher = "Association for Computational Linguistics",
1070
+ url = "https://aclanthology.org/2020.inlg-1.20",
1071
+ doi = "10.18653/v1/2020.inlg-1.20",
1072
+ pages = "138--147",
1073
+ }
1074
+
1075
+ @inproceedings{ChartQA,
1076
+ title = "{C}hart{QA}: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning",
1077
+ author = "Masry, Ahmed and
1078
+ Long, Do and
1079
+ Tan, Jia Qing and
1080
+ Joty, Shafiq and
1081
+ Hoque, Enamul",
1082
+ booktitle = "Findings of the Association for Computational Linguistics: ACL 2022",
1083
+ month = may,
1084
+ year = "2022",
1085
+ address = "Dublin, Ireland",
1086
+ publisher = "Association for Computational Linguistics",
1087
+ url = "https://aclanthology.org/2022.findings-acl.177",
1088
+ doi = "10.18653/v1/2022.findings-acl.177",
1089
+ pages = "2263--2279",
1090
+ }
1091
+
1092
+ @misc{CLEVR-Math,
1093
+ doi = {10.48550/ARXIV.2208.05358},
1094
+ url = {https://arxiv.org/abs/2208.05358},
1095
+ author = {Lindström, Adam Dahlgren},
1096
+ keywords = {Machine Learning (cs.LG), Computation and Language (cs.CL), Computer Vision and Pattern Recognition (cs.CV), FOS: Computer and information sciences, FOS: Computer and information sciences, I.2.7; I.2.10; I.2.6; I.4.8; I.1.4},
1097
+ title = {CLEVR-Math: A Dataset for Compositional Language, Visual, and Mathematical Reasoning},
1098
+ publisher = {arXiv},
1099
+ year = {2022},
1100
+ copyright = {Creative Commons Attribution Share Alike 4.0 International}
1101
+ }
1102
+
1103
+ @misc{CLEVR,
1104
+ title={CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning},
1105
+ author={Justin Johnson and Bharath Hariharan and Laurens van der Maaten and Li Fei-Fei and C. Lawrence Zitnick and Ross Girshick},
1106
+ year={2016},
1107
+ eprint={1612.06890},
1108
+ archivePrefix={arXiv},
1109
+ primaryClass={cs.CV}
1110
+ }
1111
+
1112
+ @inproceedings{CocoQA,
1113
+ author = {Ren, Mengye and Kiros, Ryan and Zemel, Richard},
1114
+ booktitle = {Advances in Neural Information Processing Systems},
1115
+ editor = {C. Cortes and N. Lawrence and D. Lee and M. Sugiyama and R. Garnett},
1116
+ pages = {},
1117
+ publisher = {Curran Associates, Inc.},
1118
+ title = {Exploring Models and Data for Image Question Answering},
1119
+ url = {https://proceedings.neurips.cc/paper_files/paper/2015/file/831c2f88a604a07ca94314b56a4921b8-Paper.pdf},
1120
+ volume = {28},
1121
+ year = {2015}
1122
+ }
1123
+
1124
+ @misc{DaTikz,
1125
+ title={AutomaTikZ: Text-Guided Synthesis of Scientific Vector Graphics with TikZ},
1126
+ author={Jonas Belouadi and Anne Lauscher and Steffen Eger},
1127
+ year={2024},
1128
+ eprint={2310.00367},
1129
+ archivePrefix={arXiv},
1130
+ primaryClass={cs.CL}
1131
+ }
1132
+
1133
+ Diagram image to text: https://huggingface.co/datasets/Kamizuru00/diagram_image_to_text by @Kamizuru00
1134
+
1135
+ @INPROCEEDINGS{DocVQA,
1136
+ author={Mathew, Minesh and Karatzas, Dimosthenis and Jawahar, C. V.},
1137
+ booktitle={2021 IEEE Winter Conference on Applications of Computer Vision (WACV)},
1138
+ title={DocVQA: A Dataset for VQA on Document Images},
1139
+ year={2021},
1140
+ volume={},
1141
+ number={},
1142
+ pages={2199-2208},
1143
+ keywords={Visualization;Computer vision;Text analysis;Image recognition;Image analysis;Conferences;Layout},
1144
+ doi={10.1109/WACV48630.2021.00225}}
1145
+
1146
+ @inproceedings{DVQA,
1147
+ title={DVQA: Understanding Data Visualizations via Question Answering},
1148
+ author={Kafle, Kushal and Cohen, Scott and Price, Brian and Kanan, Christopher},
1149
+ booktitle={CVPR},
1150
+ year={2018}
1151
+ }
1152
+
1153
+ @misc{FigureQA,
1154
+ title={FigureQA: An Annotated Figure Dataset for Visual Reasoning},
1155
+ author={Samira Ebrahimi Kahou and Vincent Michalski and Adam Atkinson and Akos Kadar and Adam Trischler and Yoshua Bengio},
1156
+ year={2018},
1157
+ eprint={1710.07300},
1158
+ archivePrefix={arXiv},
1159
+ primaryClass={cs.CV}
1160
+ }
1161
+
1162
+ @inproceedings{FinQA,
1163
+ title = "{F}in{QA}: A Dataset of Numerical Reasoning over Financial Data",
1164
+ author = "Chen, Zhiyu and
1165
+ Chen, Wenhu and
1166
+ Smiley, Charese and
1167
+ Shah, Sameena and
1168
+ Borova, Iana and
1169
+ Langdon, Dylan and
1170
+ Moussa, Reema and
1171
+ Beane, Matt and
1172
+ Huang, Ting-Hao and
1173
+ Routledge, Bryan and
1174
+ Wang, William Yang",
1175
+ editor = "Moens, Marie-Francine and
1176
+ Huang, Xuanjing and
1177
+ Specia, Lucia and
1178
+ Yih, Scott Wen-tau",
1179
+ booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
1180
+ month = nov,
1181
+ year = "2021",
1182
+ address = "Online and Punta Cana, Dominican Republic",
1183
+ publisher = "Association for Computational Linguistics",
1184
+ url = "https://aclanthology.org/2021.emnlp-main.300",
1185
+ doi = "10.18653/v1/2021.emnlp-main.300",
1186
+ pages = "3697--3711",
1187
+ }
1188
+
1189
+ @misc{GeomVerse,
1190
+ title={GeomVerse: A Systematic Evaluation of Large Models for Geometric Reasoning},
1191
+ author={Mehran Kazemi and Hamidreza Alvari and Ankit Anand and Jialin Wu and Xi Chen and Radu Soricut},
1192
+ year={2023},
1193
+ eprint={2312.12241},
1194
+ archivePrefix={arXiv},
1195
+ primaryClass={cs.CV}
1196
+ }
1197
+
1198
+ @inproceedings{hatefulmeme,
1199
+ author = {Kiela, Douwe and Firooz, Hamed and Mohan, Aravind and Goswami, Vedanuj and Singh, Amanpreet and Ringshia, Pratik and Testuggine, Davide},
1200
+ booktitle = {Advances in Neural Information Processing Systems},
1201
+ editor = {H. Larochelle and M. Ranzato and R. Hadsell and M.F. Balcan and H. Lin},
1202
+ pages = {2611--2624},
1203
+ publisher = {Curran Associates, Inc.},
1204
+ title = {The Hateful Memes Challenge: Detecting Hate Speech in Multimodal Memes},
1205
+ url = {https://proceedings.neurips.cc/paper_files/paper/2020/file/1b84c4cee2b8b3d823b30e2d604b1878-Paper.pdf},
1206
+ volume = {33},
1207
+ year = {2020}
1208
+ }
1209
+
1210
+ @inproceedings{Hitab,
1211
+ title = "{H}i{T}ab: A Hierarchical Table Dataset for Question Answering and Natural Language Generation",
1212
+ author = "Cheng, Zhoujun and
1213
+ Dong, Haoyu and
1214
+ Wang, Zhiruo and
1215
+ Jia, Ran and
1216
+ Guo, Jiaqi and
1217
+ Gao, Yan and
1218
+ Han, Shi and
1219
+ Lou, Jian-Guang and
1220
+ Zhang, Dongmei",
1221
+ editor = "Muresan, Smaranda and
1222
+ Nakov, Preslav and
1223
+ Villavicencio, Aline",
1224
+ booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
1225
+ month = may,
1226
+ year = "2022",
1227
+ address = "Dublin, Ireland",
1228
+ publisher = "Association for Computational Linguistics",
1229
+ url = "https://aclanthology.org/2022.acl-long.78",
1230
+ doi = "10.18653/v1/2022.acl-long.78",
1231
+ pages = "1094--1110",
1232
+ }
1233
+
1234
+ @article{IAM,
1235
+ author = {Marti, Urs-Viktor and Bunke, H.},
1236
+ year = {2002},
1237
+ month = {11},
1238
+ pages = {39-46},
1239
+ title = {The IAM-database: An English sentence database for offline handwriting recognition},
1240
+ volume = {5},
1241
+ journal = {International Journal on Document Analysis and Recognition},
1242
+ doi = {10.1007/s100320200071}
1243
+ }
1244
+
1245
+ @inproceedings{IconQA,
1246
+ title = {IconQA: A New Benchmark for Abstract Diagram Understanding and Visual Language Reasoning},
1247
+ author = {Lu, Pan and Qiu, Liang and Chen, Jiaqi and Xia, Tony and Zhao, Yizhou and Zhang, Wei and Yu, Zhou and Liang, Xiaodan and Zhu, Song-Chun},
1248
+ booktitle = {The 35th Conference on Neural Information Processing Systems (NeurIPS) Track on Datasets and Benchmarks},
1249
+ year = {2021}
1250
+ }
1251
+
1252
+ @INPROCEEDINGS{InfographicVQA,
1253
+ author={Mathew, Minesh and Bagal, Viraj and Tito, Rubèn and Karatzas, Dimosthenis and Valveny, Ernest and Jawahar, C. V.},
1254
+ booktitle={2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
1255
+ title={InfographicVQA},
1256
+ year={2022},
1257
+ volume={},
1258
+ number={},
1259
+ pages={2582-2591},
1260
+ keywords={Visualization;Computer vision;Computational modeling;Layout;Data visualization;Benchmark testing;Brain modeling;Document Analysis Datasets;Evaluation and Comparison of Vision Algorithms;Vision and Languages},
1261
+ doi={10.1109/WACV51458.2022.00264}
1262
+ }
1263
+
1264
+ @inproceedings{Inter-GPS,
1265
+ title = {Inter-GPS: Interpretable Geometry Problem Solving with Formal Language and Symbolic Reasoning},
1266
+ author = {Lu, Pan and Gong, Ran and Jiang, Shibiao and Qiu, Liang and Huang, Siyuan and Liang, Xiaodan and Zhu, Song-Chun},
1267
+ booktitle = {The Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021)},
1268
+ year = {2021}
1269
+ }
1270
+
1271
+ @misc{LocalizedNarratives,
1272
+ title={Connecting Vision and Language with Localized Narratives},
1273
+ author={Jordi Pont-Tuset and Jasper Uijlings and Soravit Changpinyo and Radu Soricut and Vittorio Ferrari},
1274
+ year={2020},
1275
+ eprint={1912.03098},
1276
+ archivePrefix={arXiv},
1277
+ primaryClass={cs.CV}
1278
+ }
1279
+
1280
+ @misc{MapQA,
1281
+ title={MapQA: A Dataset for Question Answering on Choropleth Maps},
1282
+ author={Shuaichen Chang and David Palzer and Jialin Li and Eric Fosler-Lussier and Ningchuan Xiao},
1283
+ year={2022},
1284
+ eprint={2211.08545},
1285
+ archivePrefix={arXiv},
1286
+ primaryClass={cs.CV}
1287
+ }
1288
+
1289
+ @misc{MIMIC-IT-General-Scene-Difference,
1290
+ title={MIMIC-IT: Multi-Modal In-Context Instruction Tuning},
1291
+ author={Bo Li and Yuanhan Zhang and Liangyu Chen and Jinghao Wang and Fanyi Pu and Jingkang Yang and Chunyuan Li and Ziwei Liu},
1292
+ year={2023},
1293
+ eprint={2306.05425},
1294
+ archivePrefix={arXiv},
1295
+ primaryClass={cs.CV}
1296
+ }
1297
+
1298
+ @inproceedings{Multihiertt,
1299
+ title = "{M}ulti{H}iertt: Numerical Reasoning over Multi Hierarchical Tabular and Textual Data",
1300
+ author = "Zhao, Yilun and
1301
+ Li, Yunxiang and
1302
+ Li, Chenying and
1303
+ Zhang, Rui",
1304
+ booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
1305
+ month = may,
1306
+ year = "2022",
1307
+ address = "Dublin, Ireland",
1308
+ publisher = "Association for Computational Linguistics",
1309
+ url = "https://aclanthology.org/2022.acl-long.454",
1310
+ pages = "6588--6600",
1311
+ }
1312
+
1313
+ @inproceedings{NLVR2,
1314
+ title = "A Corpus for Reasoning about Natural Language Grounded in Photographs",
1315
+ author = "Suhr, Alane and
1316
+ Zhou, Stephanie and
1317
+ Zhang, Ally and
1318
+ Zhang, Iris and
1319
+ Bai, Huajun and
1320
+ Artzi, Yoav",
1321
+ editor = "Korhonen, Anna and
1322
+ Traum, David and
1323
+ M{\`a}rquez, Llu{\'\i}s",
1324
+ booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics",
1325
+ month = jul,
1326
+ year = "2019",
1327
+ address = "Florence, Italy",
1328
+ publisher = "Association for Computational Linguistics",
1329
+ url = "https://aclanthology.org/P19-1644",
1330
+ doi = "10.18653/v1/P19-1644",
1331
+ pages = "6418--6428",
1332
+ }
1333
+
1334
+ @INPROCEEDINGS{OCR-VQA,
1335
+ author={Mishra, Anand and Shekhar, Shashank and Singh, Ajeet Kumar and Chakraborty, Anirban},
1336
+ booktitle={2019 International Conference on Document Analysis and Recognition (ICDAR)},
1337
+ title={OCR-VQA: Visual Question Answering by Reading Text in Images},
1338
+ year={2019},
1339
+ volume={},
1340
+ number={},
1341
+ pages={947-952},
1342
+ keywords={Optical character recognition software;Visualization;Task analysis;Knowledge discovery;Text analysis;Text recognition;Character recognition;Optical Character Recognition (OCR), Visual Question Answering (VQA), Document image analysis, textVQA},
1343
+ doi={10.1109/ICDAR.2019.00156}
1344
+ }
1345
+
1346
+ @InProceedings{okvqa,
1347
+ author = {Kenneth Marino and Mohammad Rastegari and Ali Farhadi and Roozbeh Mottaghi},
1348
+ title = {OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledge},
1349
+ booktitle = {Conference on Computer Vision and Pattern Recognition (CVPR)},
1350
+ year = {2019},
1351
+ }
1352
+
1353
+ @InProceedings{PlotQA,
1354
+ author = {Methani, Nitesh and Ganguly, Pritha and Khapra, Mitesh M. and Kumar, Pratyush},
1355
+ title = {PlotQA: Reasoning over Scientific Plots},
1356
+ booktitle = {The IEEE Winter Conference on Applications of Computer Vision (WACV)},
1357
+ month = {March},
1358
+ year = {2020}
1359
+ }
1360
+
1361
+ @inproceedings{RAVEN,
1362
+ title={RAVEN: A Dataset for Relational and Analogical Visual rEasoNing},
1363
+ author={Zhang, Chi and Gao, Feng and Jia, Baoxiong and Zhu, Yixin and Zhu, Song-Chun},
1364
+ booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
1365
+ year={2019}
1366
+ }
1367
+
1368
+ RenderedText: https://huggingface.co/datasets/wendlerc/RenderedText by @wendlerc
1369
+
1370
+ @inproceedings{Robut,
1371
+ title = "{R}obu{T}: A Systematic Study of Table {QA} Robustness Against Human-Annotated Adversarial Perturbations",
1372
+ author = "Zhao, Yilun and
1373
+ Zhao, Chen and
1374
+ Nan, Linyong and
1375
+ Qi, Zhenting and
1376
+ Zhang, Wenlin and
1377
+ Tang, Xiangru and
1378
+ Mi, Boyu and
1379
+ Radev, Dragomir",
1380
+ editor = "Rogers, Anna and
1381
+ Boyd-Graber, Jordan and
1382
+ Okazaki, Naoaki",
1383
+ booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
1384
+ month = jul,
1385
+ year = "2023",
1386
+ address = "Toronto, Canada",
1387
+ publisher = "Association for Computational Linguistics",
1388
+ url = "https://aclanthology.org/2023.acl-long.334",
1389
+ doi = "10.18653/v1/2023.acl-long.334",
1390
+ pages = "6064--6081",
1391
+ }
1392
+
1393
+ @inproceedings{SQA,
1394
+ title = "Search-based Neural Structured Learning for Sequential Question Answering",
1395
+ author = "Iyyer, Mohit and
1396
+ Yih, Wen-tau and
1397
+ Chang, Ming-Wei",
1398
+ editor = "Barzilay, Regina and
1399
+ Kan, Min-Yen",
1400
+ booktitle = "Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
1401
+ month = jul,
1402
+ year = "2017",
1403
+ address = "Vancouver, Canada",
1404
+ publisher = "Association for Computational Linguistics",
1405
+ url = "https://aclanthology.org/P17-1167",
1406
+ doi = "10.18653/v1/P17-1167",
1407
+ pages = "1821--1831",
1408
+ }
1409
+
1410
+ @misc{WikiSQL,
1411
+ title={Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning},
1412
+ author={Victor Zhong and Caiming Xiong and Richard Socher},
1413
+ year={2017},
1414
+ eprint={1709.00103},
1415
+ archivePrefix={arXiv},
1416
+ primaryClass={cs.CL}
1417
+ }
1418
+
1419
+ @inproceedings{WTQ,
1420
+ title = "Compositional Semantic Parsing on Semi-Structured Tables",
1421
+ author = "Pasupat, Panupong and
1422
+ Liang, Percy",
1423
+ editor = "Zong, Chengqing and
1424
+ Strube, Michael",
1425
+ booktitle = "Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
1426
+ month = jul,
1427
+ year = "2015",
1428
+ address = "Beijing, China",
1429
+ publisher = "Association for Computational Linguistics",
1430
+ url = "https://aclanthology.org/P15-1142",
1431
+ doi = "10.3115/v1/P15-1142",
1432
+ pages = "1470--1480",
1433
+ }
1434
+
1435
+ @inproceedings{ScienceQA,
1436
+ author = {Lu, Pan and Mishra, Swaroop and Xia, Tanglin and Qiu, Liang and Chang, Kai-Wei and Zhu, Song-Chun and Tafjord, Oyvind and Clark, Peter and Kalyan, Ashwin},
1437
+ booktitle = {Advances in Neural Information Processing Systems},
1438
+ editor = {S. Koyejo and S. Mohamed and A. Agarwal and D. Belgrave and K. Cho and A. Oh},
1439
+ pages = {2507--2521},
1440
+ publisher = {Curran Associates, Inc.},
1441
+ title = {Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering},
1442
+ url = {https://proceedings.neurips.cc/paper_files/paper/2022/file/11332b6b6cf4485b84afadb1352d3a9a-Paper-Conference.pdf},
1443
+ volume = {35},
1444
+ year = {2022}
1445
+ }
1446
+
1447
+ @inproceedings{screen2words,
1448
+ author = {Wang, Bryan and Li, Gang and Zhou, Xin and Chen, Zhourong and Grossman, Tovi and Li, Yang},
1449
+ title = {Screen2Words: Automatic Mobile UI Summarization with Multimodal Learning},
1450
+ year = {2021},
1451
+ isbn = {9781450386357},
1452
+ publisher = {Association for Computing Machinery},
1453
+ address = {New York, NY, USA},
1454
+ url = {https://doi.org/10.1145/3472749.3474765},
1455
+ doi = {10.1145/3472749.3474765},
1456
+ booktitle = {The 34th Annual ACM Symposium on User Interface Software and Technology},
1457
+ pages = {498–510},
1458
+ numpages = {13},
1459
+ keywords = {Mobile UI summarization, dataset., deep learning, language-based UI, screen understanding},
1460
+ location = {Virtual Event, USA},
1461
+ series = {UIST '21}
1462
+ }
1463
+
1464
+ @inproceedings{SpotTheDiff,
1465
+ title = "Learning to Describe Differences Between Pairs of Similar Images",
1466
+ author = "Jhamtani, Harsh and
1467
+ others",
1468
+ editor = "Riloff, Ellen and
1469
+ Chiang, David and
1470
+ Hockenmaier, Julia and
1471
+ Tsujii, Jun{'}ichi",
1472
+ booktitle = "Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing",
1473
+ month = oct # "-" # nov,
1474
+ year = "2018",
1475
+ address = "Brussels, Belgium",
1476
+ publisher = "Association for Computational Linguistics",
1477
+ url = "https://aclanthology.org/D18-1436",
1478
+ doi = "10.18653/v1/D18-1436",
1479
+ pages = "4024--4034",
1480
+ }
1481
+
1482
+ @INPROCEEDINGS{STVQA,
1483
+ author={Biten, Ali Furkan and Tito, Rubèn and Mafla, Andrés and Gomez, Lluis and Rusiñol, Marçal and Jawahar, C.V. and Valveny, Ernest and Karatzas, Dimosthenis},
1484
+ booktitle={2019 IEEE/CVF International Conference on Computer Vision (ICCV)},
1485
+ title={Scene Text Visual Question Answering},
1486
+ year={2019},
1487
+ volume={},
1488
+ number={},
1489
+ pages={4290-4300},
1490
+ keywords={Visualization;Task analysis;Knowledge discovery;Text recognition;Cognition;Computer vision;Semantics},
1491
+ doi={10.1109/ICCV.2019.00439}
1492
+ }
1493
+
1494
+ @inproceedings{TabMWP,
1495
+ title={Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning},
1496
+ author={Lu, Pan and Qiu, Liang and Chang, Kai-Wei and Wu, Ying Nian and Zhu, Song-Chun and Rajpurohit, Tanmay and Clark, Peter and Kalyan, Ashwin},
1497
+ booktitle={International Conference on Learning Representations (ICLR)},
1498
+ year={2023}
1499
+ }
1500
+
1501
+ @inproceedings{TallyQA,
1502
+ title={TallyQA: Answering Complex Counting Questions},
1503
+ author={Acharya, Manoj and Kafle, Kushal and Kanan, Christopher},
1504
+ booktitle={AAAI},
1505
+ year={2019}
1506
+ }
1507
+
1508
+ @inproceedings{TAT-QA,
1509
+ title = "{TAT}-{QA}: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance",
1510
+ author = "Zhu, Fengbin and
1511
+ Lei, Wenqiang and
1512
+ Huang, Youcheng and
1513
+ Wang, Chao and
1514
+ Zhang, Shuo and
1515
+ Lv, Jiancheng and
1516
+ Feng, Fuli and
1517
+ Chua, Tat-Seng",
1518
+ booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
1519
+ month = aug,
1520
+ year = "2021",
1521
+ address = "Online",
1522
+ publisher = "Association for Computational Linguistics",
1523
+ url = "https://aclanthology.org/2021.acl-long.254",
1524
+ doi = "10.18653/v1/2021.acl-long.254",
1525
+ pages = "3277--3287"
1526
+ }
1527
+
1528
+ @misc{textcaps,
1529
+ title={TextCaps: a Dataset for Image Captioning with Reading Comprehension},
1530
+ author={Oleksii Sidorov and Ronghang Hu and Marcus Rohrbach and Amanpreet Singh},
1531
+ year={2020},
1532
+ eprint={2003.12462},
1533
+ archivePrefix={arXiv},
1534
+ primaryClass={cs.CV}
1535
+ }
1536
+
1537
+ @inproceedings{textvqa,
1538
+ title={Towards VQA Models That Can Read},
1539
+ author={Singh, Amanpreet and Natarjan, Vivek and Shah, Meet and Jiang, Yu and Chen, Xinlei and Parikh, Devi and Rohrbach, Marcus},
1540
+ booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
1541
+ pages={8317-8326},
1542
+ year={2019}
1543
+ }
1544
+
1545
+ @INPROCEEDINGS{TQA,
1546
+ author={Kembhavi, Aniruddha and Seo, Minjoon and Schwenk, Dustin and Choi, Jonghyun and Farhadi, Ali and Hajishirzi, Hannaneh},
1547
+ booktitle={2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
1548
+ title={Are You Smarter Than a Sixth Grader? Textbook Question Answering for Multimodal Machine Comprehension},
1549
+ year={2017},
1550
+ volume={},
1551
+ number={},
1552
+ pages={5376-5384},
1553
+ keywords={Knowledge discovery;Visualization;Cognition;Training;Natural languages;Computer vision},
1554
+ doi={10.1109/CVPR.2017.571}
1555
+ }
1556
+
1557
+ @inproceedings{VisText,
1558
+ title = {{VisText: A Benchmark for Semantically Rich Chart Captioning}},
1559
+ author = {Benny J. Tang AND Angie Boggust AND Arvind Satyanarayan},
1560
+ booktitle = {The Annual Meeting of the Association for Computational Linguistics (ACL)},
1561
+ year = {2023},
1562
+ url = {http://vis.csail.mit.edu/pubs/vistext}
1563
+ }
1564
+
1565
+ @InProceedings{Visual7w,
1566
+ title = {{Visual7W: Grounded Question Answering in Images}},
1567
+ author = {Yuke Zhu and Oliver Groth and Michael Bernstein and Li Fei-Fei},
1568
+ booktitle = {{IEEE Conference on Computer Vision and Pattern Recognition}},
1569
+ year = 2016,
1570
+ }
1571
+
1572
+ @inproceedings{VisualMRC,
1573
+ author = {Ryota Tanaka and
1574
+ Kyosuke Nishida and
1575
+ Sen Yoshida},
1576
+ title = {VisualMRC: Machine Reading Comprehension on Document Images},
1577
+ booktitle = {AAAI},
1578
+ year = {2021}
1579
+ }
1580
+
1581
+ @article{VQA-RAD,
1582
+ author = {Lau, Jason and Gayen, Soumya and Ben Abacha, Asma and Demner-Fushman, Dina},
1583
+ year = {2018},
1584
+ month = {11},
1585
+ pages = {180251},
1586
+ title = {A dataset of clinically generated visual questions and answers about radiology images},
1587
+ volume = {5},
1588
+ journal = {Scientific Data},
1589
+ doi = {10.1038/sdata.2018.251}
1590
+ }
1591
+
1592
+ @misc{VQAv2,
1593
+ title={Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering},
1594
+ author={Yash Goyal and Tejas Khot and Douglas Summers-Stay and Dhruv Batra and Devi Parikh},
1595
+ year={2017},
1596
+ eprint={1612.00837},
1597
+ archivePrefix={arXiv},
1598
+ primaryClass={cs.CV}
1599
+ }
1600
+
1601
+ @misc{VSR,
1602
+ title={Visual Spatial Reasoning},
1603
+ author={Fangyu Liu and Guy Emerson and Nigel Collier},
1604
+ year={2023},
1605
+ eprint={2205.00363},
1606
+ archivePrefix={arXiv},
1607
+ primaryClass={cs.CL}
1608
+ }
1609
+
1610
+ @misc{WebSight,
1611
+ title={Unlocking the conversion of Web Screenshots into HTML Code with the WebSight Dataset},
1612
+ author={Hugo Laurençon and Léo Tronchon and Victor Sanh},
1613
+ year={2024},
1614
+ eprint={2403.09029},
1615
+ archivePrefix={arXiv},
1616
+ primaryClass={cs.HC}
1617
+ }
1618
+ </details>
1619
+
1620
+ ## Terms of Use
1621
+
1622
+ By using the dataset The Cauldron, you agree to comply with the original licenses of the sub-datasets it contains, as well as the dataset license (CC-BY-4.0). Additionally, if you use this dataset to train a Machine Learning model, you agree to disclose your use of the dataset when releasing the model or an ML application using the model.
1623
+
1624
+ ## Licensing Information
1625
+
1626
+ License CC-BY-4.0.