DataS^3 is a new benchmark demonstrating that deployment-specific data subsets can outperform full-pool training, with up to 51.3% accuracy gains, while current methods fall short.
efficiency-style
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DataS^3: Dataset Subset Selection for Specialization
DataS^3 is a new benchmark demonstrating that deployment-specific data subsets can outperform full-pool training, with up to 51.3% accuracy gains, while current methods fall short.