RoTaR: Efficient Row-Based Table Representation Learning via Teacher-Student Training Article Swipe
Zui Chen
,
Lei Cao
,
Samuel Madden
·
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2306.11696
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2306.11696
We propose RoTaR, a row-based table representation learning method, to address the efficiency and scalability issues faced by existing table representation learning methods. The key idea of RoTaR is to generate query-agnostic row representations that could be re-used via query-specific aggregation. In addition to the row-based architecture, we introduce several techniques: cell-aware position embedding, teacher-student training paradigm, and selective backward to improve the performance of RoTaR model.
Related Topics To Compare & Contrast
Concepts
Table (database)
Computer science
Scalability
Representation (politics)
Embedding
Key (lock)
Training (meteorology)
Theoretical computer science
Artificial intelligence
Machine learning
Data mining
Database
Physics
Meteorology
Political science
Politics
Law
Computer security
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2306.11696
- https://arxiv.org/pdf/2306.11696
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4381586798
All OpenAlex metadata
Raw OpenAlex JSON
No additional metadata available.