Selecting Informative Contexts Improves Language Model Fine-tuning

8792

Knowledge Triples

Source Entity	Relationship	Target Entity
Books Dataset Material	trainedOn	Language Model Fine Tuning Task
Information Gain Filtration Method	improves	Language Model Fine Tuning Task
Perplexity Metric	measures	Language Model Fine Tuning Task
Secondary Learner Method	uses	Information Gain Filtration Method
Validation Metric Metric	evaluates	Information Gain Filtration Method