Submodular maximization under a Gaussian model selects small benchmark subsets that outperform random selection for imputing leaderboard scores, with mutual information better than entropy at small sizes.
B.3 Budget Constraint for Entropy When benchmarks have non-uniform costs cj >0 , the cardinality constraint is replaced by P j∈A cj ≤ C
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Submodular Benchmark Selection
Submodular maximization under a Gaussian model selects small benchmark subsets that outperform random selection for imputing leaderboard scores, with mutual information better than entropy at small sizes.