Is there a reason that you didn't fine-tune BERT for solving TOEIC problems? Is it because the dataset for TOEIC problem solving is too small for fine-tuning?
Edit: You seem to have more than 7,000 examples of TOEIC blank-filling problems. This seems enough to fine-tune BERT for this task. Was there a reason that you didn't try that?
Is there a reason that you didn't fine-tune BERT for solving TOEIC problems? Is it because the dataset for TOEIC problem solving is too small for fine-tuning?
Edit: You seem to have more than 7,000 examples of TOEIC blank-filling problems. This seems enough to fine-tune BERT for this task. Was there a reason that you didn't try that?