pith. sign in

In particular, for each pair(j, k), gL jk(Zj, Zk) =ϕ jk(Z) ⊤β⋆ jk =c jk γjk UjUk, for some nonzero constantc jk depending only on the chosen contrast scaling

The ANOVA coefficient blocks β⋆ 0 ,{β ⋆ j },{β ⋆ jk } are an invertible linear reparameterization of(β0,{β j},{γ jk })

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
browse 1 citing papers

fields

cs.LG 1

years

2025 1

verdicts

UNVERDICTED 1

representative citing papers

CUBE: Contrastive Understanding by Balanced Experiments

cs.LG · 2025-09-13 · unverdicted · novelty 7.0

CUBE uses balanced factorial probes on black-box predictors to compute main effects and pairwise interactions as controlled contrasts within a user-specified explanation region on tabular tasks.

citing papers explorer

Showing 1 of 1 citing paper.

  • CUBE: Contrastive Understanding by Balanced Experiments cs.LG · 2025-09-13 · unverdicted · none · ref 4

    CUBE uses balanced factorial probes on black-box predictors to compute main effects and pairwise interactions as controlled contrasts within a user-specified explanation region on tabular tasks.