Pith. sign in

REVIEW 10 cited by

ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders

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 2301.00808 v1 pith:LVEW4VQE submitted 2023-01-02 cs.CV

ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders

classification cs.CV
keywords convnextlearningperformanceconvnetsimagenetmaskedmodelvarious
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Driven by improved architectures and better representation learning frameworks, the field of visual recognition has enjoyed rapid modernization and performance boost in the early 2020s. For example, modern ConvNets, represented by ConvNeXt, have demonstrated strong performance in various scenarios. While these models were originally designed for supervised learning with ImageNet labels, they can also potentially benefit from self-supervised learning techniques such as masked autoencoders (MAE). However, we found that simply combining these two approaches leads to subpar performance. In this paper, we propose a fully convolutional masked autoencoder framework and a new Global Response Normalization (GRN) layer that can be added to the ConvNeXt architecture to enhance inter-channel feature competition. This co-design of self-supervised learning techniques and architectural improvement results in a new model family called ConvNeXt V2, which significantly improves the performance of pure ConvNets on various recognition benchmarks, including ImageNet classification, COCO detection, and ADE20K segmentation. We also provide pre-trained ConvNeXt V2 models of various sizes, ranging from an efficient 3.7M-parameter Atto model with 76.7% top-1 accuracy on ImageNet, to a 650M Huge model that achieves a state-of-the-art 88.9% accuracy using only public training data.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 10 Pith papers

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

  1. Forged Calamity: Benchmark for Cross-Domain Synthetic Disaster Detection in the Age of Diffusion

    cs.CV 2026-06 unverdicted novelty 7.0

    Introduces Forged Calamity benchmark and shows that fine-tuned and zero-shot synthetic image detectors lose substantial accuracy on unseen generators and disaster types.

  2. Toward Calibrated, Fair, and accurate Deepfake Detection

    cs.LG 2026-06 unverdicted novelty 7.0

    Face-Feature Tuning is a label-free logit remapping method that reduces FPR/TPR gaps across groups in deepfake detection while preserving overall accuracy.

  3. AI-guided stimuli discovery and generation to optimize facial emotion perception studies in autism

    cs.AI 2026-07 conditional novelty 6.0

    Behavior-aligned ANNs prospectively select diagnostic facial expressions that enlarge autistic–neurotypical emotion-judgment gaps, and GAN-guided transforms of those faces reduce the gaps under phenotype-matched validation.

  4. Liquid Fusion of Heterogeneous Representations Towards General Salient Object Detection

    cs.CV 2026-06 unverdicted novelty 6.0

    LFNet fuses complementary representations from state space and convolutional backbones using liquid fusion and saliency-guided upsampling to reach state-of-the-art performance on five salient object detection tasks.

  5. Fourier Features Let Agents Learn High Precision Policies with Imitation Learning

    cs.LG 2026-06 unverdicted novelty 6.0

    Mapping point clouds to Fourier features improves high-precision imitation learning policies on RoboCasa, ManiSkill3, and real-robot tasks compared with Cartesian inputs.

  6. Transcoda: End-to-End Zero-Shot Optical Music Recognition via Data-Centric Synthetic Training

    cs.CV 2026-05 unverdicted novelty 6.0

    Transcoda achieves state-of-the-art zero-shot OMR with an 18.46% OMR-NED error rate on synthetic scores and 63.97% on historical Polish scans using a 59M model trained in 6 hours via synthetic data, kern normalization...

  7. RaPA: Enhancing Transferable Targeted Attacks via Random Parameter Pruning

    cs.LG 2025-04 conditional novelty 6.0

    Random parameter pruning during targeted attack optimization on surrogate models yields up to 11.7% higher average attack success rates when transferring to Transformer targets.

  8. When Does Sparse MoE Help in Vision? The Role of Backbone Compute Leverage in Sparse Routing

    cs.CV 2026-05 unverdicted novelty 5.0

    Sparse MoE vision models show positive accuracy gaps only when routing a substantial compute fraction ρ and using k≥2 experts at large scale; batch-axis dispatch is identified as a key failure mode.

  9. Cross-Modal Fusion of OCT and OCT angiography enface for Improved Diagnostics of Diabetic Retinopathy

    eess.IV 2026-07 conditional novelty 4.0

    Bidirectional cross-modal attention fusion of OCT B-scans with single-channel enface OCTA (real or diffusion-translated) consistently beats a ConvNeXt V2 OCT-only baseline for binary DR classification across two cohor...

  10. Physics-Constrained Neural Networks for Improved Short-Term Weather Forecasting: A Case Study over the South Pacific

    cs.LG 2026-06 unverdicted novelty 4.0

    Upgrades to WeatherGFT PCNNs with WENO-5 solver, unified autoregressive block, and two new neural backbones yield 8-22% lower RMSE at 1-12 h leads on WeatherBench South Pacific data while improving physical consistency.