AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Synthesising Counterfactual Explanations via Label-Conditional Gaussian Mixture Variational Autoencoders

Junqi Jiang, Francesco Leofante, Antonio Rago, Francesca Toni

ICLR 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Structured latent VAEs · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use LAPACE (latent path CEs) instead of FACE-InterpolateFACE-Interpolate · wine-RF actionabilityportion satisfying constraints: 0.74 -> 1.00Improved
use LAPACE instead of FACE-InterpolateFACE-Interpolate · heloc-NN actionabilityportion satisfying constraints: 0.82 -> 1.00Improved
use LAPACE instead of FACE-GreedyFACE-Greedy · wine-RF actionabilityportion satisfying constraints: 0.88 -> 1.00Improved
use LAPACE instead of FACE-GreedyFACE-Greedy · heloc-RF actionabilityportion satisfying constraints: 0.96 -> 1.00Improved
use LAPACE instead of FACE-GreedyFACE-Greedy · wine-NN and heloc-NN actionabilityportion satisfying constraints: 1.00 -> 1.00No clear effect
gradient-based constraint correction (LAPACE-constrained) instead of naive feature editsLAPACE-Naive · heloc and wine actionabilityportion satisfying constraints: 1.00 -> 1.00No clear effect
move CE point from First to Last along the latent pathLAPACE · robustness to model changesvalidity after retraining: 100%Improved
use centroid CE (LAPACE-Last) instead of nearest-neighbour CEsNNCE/RobXCE/DRCE · CE benchmarkplausibility (LOF): LOF below 1.0Improved
use LAPACE instead of DiCE/DRCE for diversityDiCE and DRCE · mixed-type datasetsdiversity: outperform on mixed-type, worse on continuous-onlyMixed
train classifier on L-GMVAE synthetic instead of real datarandom forest · adult (RF classifier)TSTR accuracy: 93.82% -> 81.13%Hurt
train classifier on L-GMVAE synthetic instead of real datarandom forest · heloc (RF classifier)TSTR accuracy: 73.97% -> 71.07%Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy