Pith. sign in

REVIEW 5 cited by

Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision

Not yet reviewed by Pith; the record is open.

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 2202.08360 v2 pith:2STVYYF3 submitted 2022-02-16 cs.CV cs.AIcs.CY

classification cs.CVcs.AIcs.CY
keywords imagesinformationmodelmodelssalientcapturescentricdata
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Discriminative self-supervised learning allows training models on any random group of internet images, and possibly recover salient information that helps differentiate between the images. Applied to ImageNet, this leads to object centric features that perform on par with supervised features on most object-centric downstream tasks. In this work, we question if using this ability, we can learn any salient and more representative information present in diverse unbounded set of images from across the globe. To do so, we train models on billions of random images without any data pre-processing or prior assumptions about what we want the model to learn. We scale our model size to dense 10 billion parameters to avoid underfitting on a large data size. We extensively study and validate our model performance on over 50 benchmarks including fairness, robustness to distribution shift, geographical diversity, fine grained recognition, image copy detection and many image classification datasets. The resulting model, not only captures well semantic information, it also captures information about artistic style and learns salient information such as geolocations and multilingual word embeddings based on visual content only. More importantly, we discover that such model is more robust, more fair, less harmful and less biased than supervised models or models trained on object centric datasets such as ImageNet.

Discussion (0). Continue with ORCID 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. Understanding the Effects of Distractors on Reasoning Vision-Language Models

    cs.CV 2025-11 conditional novelty 6.0 of 10

    Adding visual distractors to images lowers reasoning-VLM accuracy while leaving reasoning length essentially flat, with the accuracy loss traced to the share of reasoning-trace attributes devoted to distractors.

  2. Improving Remote Sensing Classification using Topological Data Analysis and Convolutional Neural Networks

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Fusing persistence homology descriptors with ResNet18 features reaches 99.33% on EuroSAT and 93.22% on RESISC45, improving both baselines and setting a claimed single-model EuroSAT record.

  3. Scaling Pre-training to One Hundred Billion Data for Vision Language Models

    cs.CV 2025-02 conditional novelty 6.0 of 10

    Scaling VLM pretraining from 10B to 100B image-text pairs yields saturation on standard benchmarks but large gains on cultural diversity, low-resource language retrieval, and subgroup disparity.

  4. Geographic Diversity Beats Data Volume for Cross-Domain Generalization in Zero-Label JEPA Driving World Models

    cs.CV 2026-07 conditional novelty 5.0 of 10

    Geographic diversity at 63K scenarios beats equal-scale and 3× larger single-geography training for zero-shot surprise transfer of JEPA driving world models from nuPlan to Argoverse 2.

  5. Data-Driven Self-Supervised Learning for the Discovery of Solution Singularity for Partial Differential Equations

    math.NA 2025-06 conditional novelty 5.0 of 10

    A density-filtering plus polynomial-fit pipeline detects PDE solution singularities from unlabeled adaptive mesh nodes.

Pith tools