Neural Data-to-Text Generation Based on Small Datasets: Comparing the Added Value of Two Semi-Supervised Learning Approaches on Top of a Large Language Model

Keyphrases

• Small Dataset 100%
• Neural Data 100%
• Semi-supervised Learning 100%
• Large Language Models 100%
• Data-to-text Generation 100%
• Supervised Learning Method 100%
• Language Model 50%
• Data-to-text Systems 50%
• Data Augmentation 33%
• Training Set 33%
• Output Quality 33%
• Pseudo-labeling 33%
• High-score 16%
• Text Quality 16%
• Learning Results 16%
• Set-valued Extension 16%
• Diversity Measure 16%
• Labeling Approach 16%
• Pre-trained Language Model 16%
• Output Diversity 16%
• Augmentation Approach 16%

Computer Science

• Learning Approach 100%
• Large Language Model 100%
• Semisupervised Learning 100%
• Language Modeling 83%
• Data Augmentation 33%
• Learning Result 16%