Pith. sign in

REVIEW 5 cited by

ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1911.09785 v2 pith:GGJVDBS4 submitted 2019-11-21 cs.LG cs.CVstat.ML

David Berthelot , Nicholas Carlini , Ekin D. Cubuk , Alex Kurakin , Kihyuk Sohn , Han Zhang , Colin Raffel This is my paper
classification cs.LGcs.CVstat.ML
keywords distributionaccuracyaugmentationalignmentanchoringdataremixmatchalgorithm
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

We improve the recently-proposed "MixMatch" semi-supervised learning algorithm by introducing two new techniques: distribution alignment and augmentation anchoring. Distribution alignment encourages the marginal distribution of predictions on unlabeled data to be close to the marginal distribution of ground-truth labels. Augmentation anchoring feeds multiple strongly augmented versions of an input into the model and encourages each output to be close to the prediction for a weakly-augmented version of the same input. To produce strong augmentations, we propose a variant of AutoAugment which learns the augmentation policy while the model is being trained. Our new algorithm, dubbed ReMixMatch, is significantly more data-efficient than prior work, requiring between $5\times$ and $16\times$ less data to reach the same accuracy. For example, on CIFAR-10 with 250 labeled examples we reach $93.73\%$ accuracy (compared to MixMatch's accuracy of $93.58\%$ with $4{,}000$ examples) and a median accuracy of $84.92\%$ with just four labels per class. We make our code and data open-source at https://github.com/google-research/remixmatch.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Self-Supervised Visual Representation Learning: Pretrain-Finetuning or Joint Training?

    cs.CV 2026-07 conditional novelty 6.0 of 10

    Jointly optimizing self-supervised and supervised losses often saves training time and helps in low-label settings, but pretrain-then-finetune remains better for several contrastive methods and specialized domains.

  2. Semi-MedRef: Semi-Supervised Medical Referring Image Segmentation with Cross-Modal Alignment

    cs.CV 2026-05 unverdicted novelty 6.0 of 10

    Semi-MedRef introduces T-PatchMix, PosAug, and ITCL within a teacher-student SSL setup to preserve image-text alignment under augmentation for medical referring segmentation on QaTa-COV19 and MosMedData+.

  3. JanusNet: Hierarchical Slice-Block Shuffle and Displacement for Semi-Supervised 3D Multi-Organ Segmentation

    cs.CV 2025-08 conditional novelty 6.0 of 10

    Slice-block shuffle plus confidence-guided displacement reports 72.67% average Dice on Synapse with 20% labels and 63.99% on AMOS with 5% labels.

  4. FixCLR: Negative-Class Contrastive Learning for Semi-Supervised Domain Generalization

    cs.CV 2025-06 conditional novelty 6.0 of 10

    A plug-and-play contrastive regularization term, FixCLR, repels different pseudo-classes across domains and improves semi-supervised domain generalization accuracy when combined with FixMatch-based methods.

  5. Learning from Limited and Imperfect Data

    cs.LG 2025-07 unverdicted novelty 3.0 of 10

    A doctoral thesis compiling nine peer-reviewed papers on long-tailed image generation, long-tailed recognition, semi-supervised learning, and domain adaptation.

Pith tools