MGCoT: Multi-Grained Contextual Transformer for table-based text generation

Xianjie Mo, Yang Xiang, Youcheng Pan, Yongshuai Hou, Ping Luo

Abstract

Highlights•The contexts for each token in a table are various from the structural perspective.•Forming the local contexts allows models to capture contexts in a dynamic range.•Forming the shared global context allows models to capture the consensus.•Models can flexibly and holistically comprehend a table via multi-grained contexts.
BibTeX
@article{mo2024mgcot,
  author = {Xianjie Mo and Yang Xiang and Youcheng Pan and Yongshuai Hou and Ping Luo},
  title = {{MGCoT: Multi-Grained Contextual Transformer for table-based text generation}},
  journal = {Expert Systems with Applications},
  year = {2024},
  doi = {10.1016/j.eswa.2024.123742},
}

Loading related content...

Figure