loading page

Towards Generating Financial Reports From Table Data Using Transformers
  • +3
  • Clayton Chapman ,
  • Lars Hillebrand ,
  • Marc Robin Stenzel ,
  • Tobias Deusser ,
  • Christian Bauckhage ,
  • Rafet Sifa
Clayton Chapman
Author Profile
Lars Hillebrand
Author Profile
Marc Robin Stenzel
Author Profile
Tobias Deusser
Author Profile
Christian Bauckhage
Author Profile
Rafet Sifa
Author Profile


Financial reports are commonplace in the business world, but are long and tedious to produce. These reports mostly consist of tables with written sections describing these tables. Automating the process of creating these reports, even partially has the potential to save a company time and resources that could be spent on more creative tasks. We implement a transformer network to solve the task of generating this text. By generating matching pairs between tables and sentences found in financial documents, we created a dataset for our transformer. We were able to achieve promising results, with the final model reaching a BLEU score of 63.3. Generated sentences are natural, grammatically correct and mostly faithful to the information found in the tables.