Improving User Controlled Table-To-Text Generation Robustness
This addresses robustness issues in user-controlled table-to-text generation for practical applications, but it is incremental as it builds on existing fine-tuning methods.
The paper tackled the problem of table-to-text generation models performing poorly on realistic noisy user cell selections, and found that fine-tuning with simulated noisy inputs improved BLEU scores by 4.85 points on noisy cases and 1.4 on clean cases, achieving state-of-the-art performance on the ToTTo dataset.
In this work we study user controlled table-to-text generation where users explore the content in a table by selecting cells and reading a natural language description thereof automatically produce by a natural language generator. Such generation models usually learn from carefully selected cell combinations (clean cell selections); however, in practice users may select unexpected, redundant, or incoherent cell combinations (noisy cell selections). In experiments, we find that models perform well on test sets coming from the same distribution as the train data but their performance drops when evaluated on realistic noisy user inputs. We propose a fine-tuning regime with additional user-simulated noisy cell selections. Models fine-tuned with the proposed regime gain 4.85 BLEU points on user noisy test cases and 1.4 on clean test cases; and achieve comparable state-of-the-art performance on the ToTTo dataset.