A base-sized pre-trained model designed to enhance table-based question answering and fact verification tasks.
The TAPEX Base Model is a foundational model pre-trained to empower existing systems with advanced table reasoning capabilities. Built upon the BART architecture, it utilizes a transformer encoder-decoder framework with a bidirectional encoder and an autoregressive decoder. This model is particularly adept at understanding and processing tabular data, making it suitable for tasks such as table-based question answering and fact verification.
MIT
Qian Liu and Bei Chen and Jiaqi Guo and Morteza Ziyadi and Zeqi Lin and Weizhu Chen and Jian-Guang Lou
Fine-Tuned Model
N.A.
Open
Sector Agnostic
12/03/25 06:34:46
0
MIT
© 2026 - Copyright AIKosh. All rights reserved. This portal is developed by National e-Governance Division for AIKosh mission.