LAMOL

Reported on 2 benchmarks across 1 task · 1 paper · 1 SOTA

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

Methodology2 results

Continual LearningonASC (19 tasks)
F1 - macro· 2019-09-07
0.8059
best: 0.8811 (Multi-task Learning (MTL; Upper Bound))
SOTA
LAMOL: LAnguage MOdeling for Lifelong Language Learning arXiv:1909.03329
Continual Learningon20Newsgroup (10 tasks)
F1 - macro· 2019-09-07
0.4572
best: 0.9523 (CTR)
LAMOL: LAnguage MOdeling for Lifelong Language Learning arXiv:1909.03329