B-CL

Reported on 3 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology3 results

Continual Learningon20Newsgroup (10 tasks)
F1 - macro· 2021-12-06
0.9504
best: 0.9523 (CTR)
Adapting BERT for Continual Learning of a Sequence of Aspect Sentiment Classification Tasks arXiv:2112.03271
Continual LearningonASC (19 tasks)
F1 - macro· 2021-12-06
0.814
best: 0.8811 (Multi-task Learning (MTL; Upper Bound))
Adapting BERT for Continual Learning of a Sequence of Aspect Sentiment Classification Tasks arXiv:2112.03271
Continual LearningonDSC (10 tasks)
F1 - macro· 2021-12-06
0.7651
best: 0.8875 (CTR)
Adapting BERT for Continual Learning of a Sequence of Aspect Sentiment Classification Tasks arXiv:2112.03271