AI Paper Atlas

Research map / Learning theory and algorithms / Optimal transport

Wasserstein Distances Made Explainable: Insights Into Dataset Shifts and Transport Phenomena

Philip Naumann, Jacob Kauffmann, Grégoire Montavon

TPAMI 2026 · Learning theory and algorithms › Optimal transport · approach: Sliced Wasserstein variants · 7 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 WaX instead of Occlusion at high qWasserstein explanation · Robot (W2, q=inf)SRG: 0.72 -> 1.14Improved
use WaX instead of MeanShift at high pWasserstein explanation · Robot (W10, q=2)SRG: 4.81 -> 8.33Improved
add alpha and beta tuning to Coupling (WaX instead of Coupling) (ablation)Coupling baseline · Robot (W10, q=2)SRG: -0.68 -> 8.33Improved
use WaX instead of Coupling at q=infCoupling baseline · Crime (W2, q=inf)SRG: 0.25 -> 0.59Improved
use WaX instead of Occlusion at low q (equal accuracy, cheaper)Occlusion · Robot (W1, q=1)SRG: 9.10 -> 9.11No clear effect
use WaX instead of MeanShift on Robot W1MeanShift · Robot (W1, q=2)SRG: 0.06 -> 1.48Improved
use exact coupling instead of uniform coupling in WaX (ablation)WaX · Air Quality (every 1h)cosine similarity: 0.80 -> 0.85Improved
use uniform coupling instead of exact coupling in WaX (ablation)WaX · Electricity (every 1h)cosine similarity: 0.83 -> 0.88Improved
use WaX instead of Logistic [GI] baselineLogistic · Air Quality (every 1h)cosine similarity: 0.47 -> 0.85Improved
use WaX instead of Uniform relevanceUniform · Appliances (every 1h)cosine similarity: 0.38 -> 0.51Improved
use WaX instead of Uniform on ElectricityUniform · Electricity (every 6h)cosine similarity: 0.94 -> 0.98No clear effect
use WaX exact coupling instead of MeanShift on PLISM UNIMeanShift · PLISM (UNI)cosine similarity: 0.77 -> 0.99Improved

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