mlsanityData to 7 Oct 2026

Learning Robust Global Representations by Penalizing Local Predictive Power

arXiv 1905.13549 ↗First posted 29 May 2019cs.CVcited by 1,010

Not on the map. Why?

Cited by 339 in Multimodal LLMs / vision-language models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Despite their renowned predictive power on i.i.d. data, convolutional neural networks are known to rely more on high-frequency patterns that humans deem superficial than on low-frequency patterns that agree better with intuitions about what constitutes category membership. This paper proposes a method for training robust convolutional networks by penalizing the predictive power of the local representations learned by earlier layers. Intuitively, our networks are forced to discard predictive signals such as color and texture that can be gleaned from local receptive fields and to rely instead on the global structures of the image. Across a battery of synthetic and benchmark domain adaptation tasks, our method confers improved generalization out of the domain. Also, to evaluate cross-domain transfer, we introduce ImageNet-Sketch, a new dataset consisting of sketch-like images, that matches the ImageNet classification validation set in categories and scale.

Key facts. Cited by 1,010 papers in the map (data through 7 Oct 2026). First posted on arXiv 29 May 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 52; 2022: 135; 2023: 202; 2024: 266; 2025: 226; 2026: 129.

1,010 papers.

Sort:

By area:

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  3. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  4. 2022A ConvNet for the 2020scited by 3,545
  5. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,534
  6. 2021Learning to Prompt for Vision-Language Modelscited by 2,269
  7. 2022Conditional Prompt Learning for Vision-Language Modelscited by 1,427
  8. 2022Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timecited by 1,366
  9. 2022CoCa: Contrastive Captioners are Image-Text Foundation Modelscited by 1,265
  10. 2022Reproducible scaling laws for contrastive language-image learningcited by 1,163
  11. 2021CLIP-Adapter: Better Vision-Language Models with Feature Adapterscited by 1,021
  12. 2023EVA-CLIP: Improved Training Techniques for CLIP at Scalecited by 786
38 more
  1. 2021Robust fine-tuning of zero-shot modelscited by 785
  2. 2022PaLI: A Jointly-Scaled Multilingual Language-Image Modelcited by 762
  3. 2022Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distributioncited by 722
  4. 2022MaPLe: Multi-modal Prompt Learningcited by 685
  5. 2022EVA: Exploring the Limits of Masked Visual Representation Learning at Scalecited by 685
  6. 2023DataComp: In search of the next generation of multimodal datasetscited by 635
  7. 2021Merging Models with Fisher-Weighted Averagingcited by 613
  8. 2021LiT: Zero-Shot Transfer with Locked-image text Tuningcited by 572
  9. 2022Beyond neural scaling laws: beating power law scaling via data pruningcited by 519
  10. 2023Symbolic Discovery of Optimization Algorithmscited by 423
  11. 2022Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Modelscited by 379
  12. 2023EVA-02: A Visual Representation for Neon Genesiscited by 377
  13. 2022Prompt-aligned Gradient for Prompt Tuningcited by 339
  14. 2022Is synthetic data from generative models ready for image recognition?cited by 320
  15. 2025Perception Encoder: The best visual embeddings are not at the output of the networkcited by 315
  16. 2023Visual-Language Prompt Tuning with Knowledge-guided Context Optimizationcited by 307
  17. 2022Masked Siamese Networks for Label-Efficient Learningcited by 301
  18. 2022BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizerscited by 298
  19. 2023SemDeDup: Data-efficient learning at web-scale through semantic deduplicationcited by 295
  20. 2022MultiMAE: Multi-modal Multi-task Masked Autoencoderscited by 287
  21. 2023Self-regulating Prompts: Foundational Model Adaptation without Forgettingcited by 272
  22. 2022Matryoshka Representation Learningcited by 268
  23. 2022What does a platypus look like? Generating customized prompts for zero-shot image classificationcited by 263
  24. 2021Gradient Matching for Domain Generalizationcited by 258
  25. 2023PaLI-X: On Scaling up a Multilingual Vision and Language Modelcited by 254
  26. 2023A Cookbook of Self-Supervised Learningcited by 248
  27. 2023Sequential Modeling Enables Scalable Learning for Large Vision Modelscited by 212
  28. 2021Partial success in closing the gap between human and machine visioncited by 185
  29. 2021Combined Scaling for Zero-shot Transfer Learningcited by 182
  30. 2022Fake it till you make it: Learning transferable representations from synthetic ImageNet clonescited by 171
  31. 2021Learning to Diversify for Single Domain Generalizationcited by 170
  32. 2023FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterizationcited by 161
  33. 2024SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Accelerationcited by 158
  34. 2023Diverse Data Augmentation with Diffusions for Effective Test-time Prompt Tuningcited by 150
  35. 2022Finetune like you pretrain: Improved finetuning of zero-shot vision modelscited by 149
  36. 2022Unified Vision and Language Prompt Learningcited by 145
  37. 2022Leveraging Unlabeled Data to Predict Out-of-Distribution Performancecited by 141
  38. 2024Efficient Test-Time Adaptation of Vision-Language Modelscited by 138

50 most cited of 1,010 All 1,010 (JSON)