ABSTRACT
Tables are widely used for data storage and presentation due to their high flexibility in layout. The importance of tables as information carriers and the complexity of tabular data understanding attract a great deal of research on large-scale pre-training for tabular data. However, most of the works design models for specific types of tables, such as relational tables and tables with well-structured headers, neglecting tables with complex layouts. In real-world scenarios, there are many such tables beyond their target scope that cannot be well supported. In this paper, we propose GetPt, a unified pre-training architecture for general table representation applicable even to tables with complex structures and layouts. First, we convert a table to a heterogeneous graph with multiple types of edges to represent the layout of the table. Based on the graph, a specially designed transformer is applied to jointly model the semantics and structure of the table. Second, we devise the Alternate Attention Network (AAN) to better model the contextual information across multiple granularities of a table including tokens, cells, and the table. To better support a wide range of downstream tasks, we further employ three pre-training objectives and pre-train the model on a large table dataset. We fine-tune and evaluate GetPt model on two representative tasks, table type classification, and table structure recognition. Experiments show that GetPt outperforms existing state-of-the-art methods on these tasks.
Supplemental Material
- Chandra Sekhar Bhagavatula, Thanapon Noraset, and Doug Downey. 2015. Tabel: Entity linking in web tables. In International Semantic Web Conference. Springer, 425--441.Google ScholarDigital Library
- Zhoujun Cheng, Haoyu Dong, Ran Jia, Pengfei Wu, Shi Han, Fan Cheng, and Dongmei Zhang. 2021. FORTAP: Using Formulas for Numerical-Reasoning-Aware Table Pretraining. arXiv preprint arXiv:2109.07323 (2021).Google Scholar
- Kevin Clark, Minh-Thang Luong, Quoc V Le, and Christopher D Manning. 2020. Electra: Pre-training text encoders as discriminators rather than generators. arXiv preprint arXiv:2003.10555 (2020).Google Scholar
- Eric Crestan and Patrick Pantel. 2011. Web-scale table census and classification. In Proceedings of the fourth ACM international conference on Web search and data mining. 545--554.Google ScholarDigital Library
- Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. https://arxiv.org/abs/1810.04805Google Scholar
- Haoyu Dong, Zhoujun Cheng, Xinyi He, Mengyu Zhou, Anda Zhou, Fan Zhou, Ao Liu, Shi Han, and Dongmei Zhang. 2022. Table pretraining: A survey on model architectures, pretraining objectives, and downstream tasks. arXiv preprint arXiv:2201.09745 (2022).Google Scholar
- Haoyu Dong, Shijie Liu, Zhouyu Fu, Shi Han, and Dongmei Zhang. 2019a. Semantic Structure Extraction for Spreadsheet Tables with a Multi-task Learning Architecture.Google Scholar
- Haoyu Dong, Shijie Liu, Zhouyu Fu, Shi Han, and Dongmei Zhang. 2019b. Semantic structure extraction for spreadsheet tables with a multi-task learning architecture. In Workshop on Document Intelligence at NeurIPS 2019.Google Scholar
- Haoyu Dong, Jiong Yang, Shi Han, and Dongmei Zhang. 2020. Learning formatting style transfer and structure extraction for spreadsheet tables with a hybrid neural network architecture. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management. 2389--2396.Google ScholarDigital Library
- Lun Du, Fei Gao, Xu Chen, Ran Jia, Junshan Wang, Jiang Zhang, Shi Han, and Dongmei Zhang. 2021. TabularNet: A neural network architecture for understanding semantic structures of tabular data. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 322--331.Google ScholarDigital Library
- Vijay Prakash Dwivedi and Xavier Bresson. 2020. A generalization of transformer networks to graphs. arXiv preprint arXiv:2012.09699 (2020).Google Scholar
- Julian Eberius, Katrin Braunschweig, Markus Hentsch, Maik Thiele, Ahmad Ahmadov, and Wolfgang Lehner. 2015a. Building the Dresden Web Table Corpus: A Classification Approach. 2015 IEEE/ACM 2nd International Symposium on Big Data Computing (BDC) (2015), 41--50.Google Scholar
- Julian Eberius, Katrin Braunschweig, Markus Hentsch, Maik Thiele, Ahmad Ahmadov, and Wolfgang Lehner. 2015b. Building the dresden web table corpus: A classification approach. In 2015 IEEE/ACM 2nd International Symposium on Big Data Computing (BDC). IEEE, 41--50.Google ScholarCross Ref
- Julian Martin Eisenschlos, Maharshi Gor, Thomas Müller, and William W. Cohen. 2021. MATE: Multi-view Attention for Table Transformer Efficiency. In Conference on Empirical Methods in Natural Language Processing.Google Scholar
- Majid Ghasemi-Gol, Jay Pujara, and Pedro A. Szekely. 2019. Tabular Cell Classification Using Pre-Trained Cell Embeddings. 2019 IEEE International Conference on Data Mining (ICDM) (2019), 230--239.Google Scholar
- Majid Ghasemi-Gol and Pedro Szekely. 2018. TabVec: Table Vectors for Classification of Web Tables. https://arxiv.org/abs/1802.06290Google Scholar
- Jonathan Herzig, Pawel Krzysztof Nowak, Thomas Müller, Francesco Piccinno, and Julian Eisenschlos. 2020. TaPas: Weakly Supervised Table Parsing via Pre-training. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Association for Computational Linguistics. https://doi.org/10.18653%2Fv1%2F2020.acl-main.398Google ScholarCross Ref
- Nathan Hurst, Kim Marriott, and Peter Moulder. 2005. Toward tighter tables. In Proceedings of the 2005 ACM symposium on Document engineering. 74--83.Google ScholarDigital Library
- Hiroshi Iida, Dung Thai, Varun Manjunatha, and Mohit Iyyer. 2021. TABBIE: Pretrained Representations of Tabular Data. https://arxiv.org/abs/2105.02584Google Scholar
- Thomas N Kipf and Max Welling. 2016. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907 (2016).Google Scholar
- Elvis Koci, Maik Thiele, Josephine Rehak, Oscar Romero, and Wolfgang Lehner. 2019. DECO: A Dataset of Annotated Spreadsheets for Layout and Table Recognition. 2019 International Conference on Document Analysis and Recognition (ICDAR) (2019), 1280--1285.Google Scholar
- Guillaume Lample and Alexis Conneau. 2019. Cross-lingual Language Model Pretraining. https://arxiv.org/abs/1901.07291Google Scholar
- Larissa R Lautert, Marcelo M Scheidt, and Carina F Dorneles. 2013. Web table taxonomy and formalization. ACM SIGMOD Record, Vol. 42, 3 (2013), 28--33.Google ScholarDigital Library
- Weihong Lin, Zheng Sun, Chixiang Ma, Mingze Li, Jiawei Wang, Lei Sun, and Qiang Huo. 2022. TSRFormer: Table Structure Recognition with Transformers. In Proceedings of the 30th ACM International Conference on Multimedia. 6473--6482.Google ScholarDigital Library
- Xiaofan Lin. 2006. Active layout engine: Algorithms and applications in variable data printing. Computer-Aided Design, Vol. 38, 5 (2006), 444--456.Google ScholarDigital Library
- Hao Liu, Xin Li, Bing Liu, Deqiang Jiang, Yinsong Liu, and Bo Ren. 2022. Neural collaborative graph machines for table structure recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 4533--4542.Google ScholarCross Ref
- Qian Liu, Bei Chen, Jiaqi Guo, Zeqi Lin, and Jian-guang Lou. 2021. Tapex: Table pre-training via learning a neural sql executor. arXiv preprint arXiv:2107.07653 (2021).Google Scholar
- Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692 (2019).Google Scholar
- Ilya Loshchilov and Frank Hutter. 2017. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101 (2017).Google Scholar
- George A Miller. 1995. WordNet: a lexical database for English. Commun. ACM, Vol. 38, 11 (1995), 39--41.Google ScholarDigital Library
- Kyosuke Nishida, Kugatsu Sadamitsu, Ryuichiro Higashinaka, and Yoshihiro Matsuo. 2017a. Understanding the Semantic Structures of Tables with a Hybrid Deep Neural Network Architecture. In AAAI.Google Scholar
- Kyosuke Nishida, Kugatsu Sadamitsu, Ryuichiro Higashinaka, and Yoshihiro Matsuo. 2017b. Understanding the semantic structures of tables with a hybrid deep neural network architecture. In Thirty-First AAAI Conference on Artificial Intelligence.Google ScholarDigital Library
- Shah Rukh Qasim, Hassan Mahmood, and Faisal Shafait. 2019. Rethinking table recognition using graph neural networks. In 2019 International Conference on Document Analysis and Recognition (ICDAR). IEEE, 142--147.Google ScholarCross Ref
- Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf. 2019. DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. https://arxiv.org/abs/1910.01108Google Scholar
- Kexuan Sun, Harsha Rayudu, and Jay Pujara. 2021. A Hybrid Probabilistic Approach for Table Understanding. In AAAI Conference on Artificial Intelligence.Google ScholarCross Ref
- Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention is All You Need. https://arxiv.org/pdf/1706.03762.pdfGoogle Scholar
- Petar Velivc kovi? , Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, and Yoshua Bengio. 2017. Graph attention networks. arXiv preprint arXiv:1710.10903 (2017).Google Scholar
- Daheng Wang, Prashant Shiralkar, Colin Lockard, Binxuan Huang, Xin Luna Dong, and Meng Jiang. 2021b. TCN: Table Convolutional Network for Web Table Interpretation. In Proceedings of the Web Conference 2021. ACM. https://doi.org/10.1145/3442381.3450090Google ScholarDigital Library
- Fei Wang, Kexuan Sun, Muhao Chen, Jay Pujara, and Pedro Szekely. 2021c. Retrieving Complex Tables with Multi-Granular Graph Representation Learning. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. ACM. https://doi.org/10.1145%2F3404835.3462909Google ScholarDigital Library
- Haoyu Wang, Muhao Chen, Hongming Zhang, and Dan Roth. 2020. Joint constrained learning for event-event relation extraction. arXiv preprint arXiv:2010.06727 (2020).Google Scholar
- Zhiruo Wang, Haoyu Dong, Ran Jia, Jia Li, Zhiyi Fu, Shi Han, and Dongmei Zhang. 2021a. Tuta: Tree-based transformers for generally structured table pre-training. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 1780--1790.Google ScholarDigital Library
- Kan Wu, Houwen Peng, Minghao Chen, Jianlong Fu, and Hongyang Chao. 2021. Rethinking and Improving Relative Position Encoding for Vision Transformer. 2021 IEEE/CVF International Conference on Computer Vision (ICCV) (2021), 10013--10021.Google Scholar
- Wenyuan Xue, Baosheng Yu, Wen Wang, Dacheng Tao, and Qingyong Li. 2021. Tgrnet: A table graph reconstruction network for table structure recognition. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 1295--1304.Google ScholarCross Ref
- Zhilin Yang, Zihang Dai, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, and Quoc V. Le. 2019. XLNet: Generalized Autoregressive Pretraining for Language Understanding. https://arxiv.org/abs/1906.08237Google Scholar
- Pengcheng Yin, Graham Neubig, Wen-tau Yih, and Sebastian Riedel. 2020. TaBERT: Pretraining for Joint Understanding of Textual and Tabular Data. https://arxiv.org/abs/2005.08314Google Scholar
- Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, and Tie-Yan Liu. 2021. Do transformers really perform badly for graph representation? Advances in Neural Information Processing Systems , Vol. 34 (2021), 28877--28888.Google Scholar
- Guangyu Zhou, Muhao Chen, Chelsea JT Ju, Zheng Wang, Jyun-Yu Jiang, and Wei Wang. 2020. Mutation effect estimation on protein--protein interactions using deep contextualized representation learning. NAR genomics and bioinformatics , Vol. 2, 2 (2020), lqaa015. ioGoogle Scholar
Index Terms
- GetPt: Graph-enhanced General Table Pre-training with Alternate Attention Network
Recommendations
Schema dependency-enhanced curriculum pre-training for table semantic parsing
AbstractLarge pre-trained models exhibit improved table-semantic-parsing performances by leveraging large-scale corpora to enhance the representation learning ability of semantic parsers. However, existing table pre-training methods do not sufficiently ...
Automated 3D Pre-Training for Molecular Property Prediction
KDD '23: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data MiningMolecular property prediction is an important problem in drug discovery and materials science. As geometric structures have been demonstrated necessary for molecular property prediction,3D information has been combined with various graph learning ...
Poster: Boosting Adversarial Robustness by Adversarial Pre-training
CCS '23: Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications SecurityVision Transformer (ViT) shows superior performance on various tasks, but, similar to other deep learning techniques, it is vulnerable to adversarial attacks. Due to the differences between ViT and traditional CNNs, previous works designed new ...
Comments