AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Okapi: Generalising Better by Making Statistical Matches Match

Myles Bartlett, Sara Romiti, Viktoriia Sharmanska, Novi Quadrianto

NeurIPS 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervision in supervised pipelines · 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
switch backbone from ResNet to ConvNeXtERM · iWildCamOOD macro F1: 32.2 -> 33.3Improved
switch backbone from ResNet to ConvNeXtFixMatch · iWildCamOOD macro F1: 31.0 -> 35.2Improved
switch backbone from ResNet to ConvNeXtERM · PovertyMapOOD worst U/R correlation: 0.49 -> 0.53Improved
switch backbone from ResNet to ConvNeXtFixMatch · PovertyMapOOD worst U/R correlation: 0.30 -> 0.34Improved
use ConvNeXt so FixMatch surpasses ERMFixMatch · iWildCamOOD macro F1: 31.0 vs 32.2 -> 35.2 vs 33.3Improved
add Okapi cross-domain consistency lossERM (ConvNeXt) · iWildCamOOD macro F1: 33.3 -> 36.1Improved
add OkapiFixMatch (ConvNeXt) · iWildCamOOD macro F1: 35.2 -> 36.1Improved
add OkapiERM (ConvNeXt) · PovertyMapOOD worst U/R correlation: 0.53 -> 0.55Improved
add OkapiERM (ConvNeXt) · PovertyMapOOD worst U/R MSE: 0.35 -> 0.33No clear effect
FixMatch without confidence thresholding for regressionFixMatch · PovertyMapOOD worst U/R correlation: 0.53 -> 0.34Hurt
add OkapiERM (reproduction) · CivilCommentsOOD worst-group accuracy: 68.5 -> 69.7Improved
remove caliper filtering (standard kNN) (ablation)Okapi · PovertyMapOOD worst U/R correlation: 0.54 -> 0.55Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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