By turning each teacher's labels into a separate confidence-prediction task conditioned on a teacher token, Teacher2Task claims to outperform the best individual teacher on open-vocabulary tagging without manual aggregation heuristics.
Distilling knowledge from ensembles of neural networks for speech recognition,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Heuristic-Free Multi-Teacher Learning
By turning each teacher's labels into a separate confidence-prediction task conditioned on a teacher token, Teacher2Task claims to outperform the best individual teacher on open-vocabulary tagging without manual aggregation heuristics.