Introduces complement-aware submodular functions (CSI) that preserve structure between subset and complement for improved robust data selection.
Submodular Benchmark Selection
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
abstract
Evaluating large language models across many benchmarks is expensive, yet many benchmarks are highly correlated. We formalize the selection of a small, informative subset as submodular maximization under a multivariate Gaussian model. Entropy (log-determinant covariance) and mutual information between selected and remaining benchmarks arise as natural objectives. Both are submodular; entropy selection coincides with pivoted Cholesky and has spectral residual bounds, while mutual information is non-monotone in general but empirically monotone for small subsets, so we optimize it greedily. Experiments on three matrices from ten public leaderboards show that mutual information selection outperforms entropy for imputation at small subsets.
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
ProactBench measures LLM conversational proactivity in three phases using 198 multi-agent dialogues and finds recovery behavior hard to predict from existing benchmarks.
citing papers explorer
-
Complement Submodular Information Measures for Balanced and Robust Data Selection
Introduces complement-aware submodular functions (CSI) that preserve structure between subset and complement for improved robust data selection.
-
ProactBench: Beyond What The User Asked For
ProactBench measures LLM conversational proactivity in three phases using 198 multi-agent dialogues and finds recovery behavior hard to predict from existing benchmarks.