mlsanityData to 7 Oct 2026

A Data-Based Perspective on Transfer Learning

arXiv 2207.05739 ↗First posted 12 Jul 2022cs.LGcited by 32

Not on the map. Why?

Abstract (arXiv)

It is commonly believed that in transfer learning including more pre-training data translates into better performance. However, recent evidence suggests that removing data from the source dataset can actually help too. In this work, we take a closer look at the role of the source dataset's composition in transfer learning and present a framework for probing its impact on downstream performance. Our framework gives rise to new capabilities such as pinpointing transfer learning brittleness as well as detecting pathologies such as data-leakage and the presence of misleading examples in the source dataset. In particular, we demonstrate that removing detrimental datapoints identified by our framework improves transfer learning performance from ImageNet on a variety of target tasks. Code is available at https://github.com/MadryLab/data-transfer

Key facts. Cited by 32 papers in the map (data through 7 Oct 2026). Builds on 30 of its 56 references that are in the map. First posted on arXiv 12 Jul 2022.

Builds on

30 of 56 references.

References · 30

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  3. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,239
  4. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
  5. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  6. 2013Fine-Grained Visual Classification of Aircraftcited by 1,701
  7. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 969
  8. 2019Big Transfer (BiT): General Visual Representation Learningcited by 766
  9. 2022Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distributioncited by 722
  10. 2019Data Shapley: Equitable Valuation of Data for Machine Learningcited by 617
  11. 2018Do Better ImageNet Models Transfer Better?cited by 560
  12. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 492
18 more
  1. 2016R-FCN: Object Detection via Region-based Fully Convolutional Networkscited by 414
  2. 2020What is being transferred in transfer learning?cited by 365
  3. 2020Do Adversarially Robust ImageNet Models Transfer Better?cited by 315
  4. 2018SentEval: An Evaluation Toolkit for Universal Sentence Representationscited by 294
  5. 2016What makes ImageNet good for transfer learning?cited by 192
  6. 2022Datamodels: Predicting Predictions from Training Datacited by 186
  7. 2021On the Robustness of Vision Transformers to Adversarial Examplescited by 167
  8. 2021Beta Shapley: a Unified and Noise-reduced Data Valuation Framework for Machine Learningcited by 162
  9. 2016Panning for Gold: Model-X Knockoffs for High-dimensional Controlled Variable Selectioncited by 120
  10. 2018Interactive Supercomputing on 40,000 Cores for Machine Learning and Data Analysiscited by 110
  11. 2021CheXtransfer: Performance and Parameter Efficiency of ImageNet Models for Chest X-Ray Interpretationcited by 57
  12. 2019Learning What and Where to Transfercited by 43
  13. 2015Transfer Learning from Deep Features for Remote Sensing and Poverty Mappingcited by 37
  14. 2022Missingness Bias in Model Debuggingcited by 36
  15. 2018Domain Adaptive Transfer Learning with Specialist Modelscited by 31
  16. 2020Adversarially-Trained Deep Nets Transfer Better: Illustration on Image Classificationcited by 18
  17. 2022Data Debugging with Shapley Importance over End-to-End Machine Learning Pipelinescited by 16
  18. 2019Self-Driving Car Steering Angle Prediction Based on Image Recognitioncited by 14

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 9; 2024: 8; 2025: 10; 2026: 1.

32 papers.

Sort:

By area:

  1. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generationcited by 376
  2. 2023Model Sparsity Can Simplify Machine Unlearningcited by 206
  3. 2022Understanding and Improving Visual Prompting: A Label-Mapping Perspectivecited by 86
  4. 2022Discovering Bugs in Vision Models using Off-the-shelf Image Generation and Captioningcited by 45
  5. 2024IRSAM: Advancing Segment Anything Model for Infrared Small Target Detectioncited by 45
  6. 2023Know Your Self-supervised Learning: A Survey on Image-based Generative and Discriminative Trainingcited by 39
  7. 2024Unleashing the Power of Data Tsunami: A Comprehensive Survey on Data Assessment and Selection for Instruction Tuning of Language Modelscited by 29
  8. 2025DataMIL: Selecting Data for Robot Imitation Learning with Datamodelscited by 25
  9. 2023Selective Pre-training for Private Fine-tuningcited by 22
  10. 2024Score Forgetting Distillation: A Swift, Data-Free Method for Machine Unlearning in Diffusion Modelscited by 22
  11. 2023On the Trade-off of Intra-/Inter-class Diversity for Supervised Pre-trainingcited by 14
  12. 2022The Value of Out-of-Distribution Datacited by 12
20 more
  1. 2023Selectivity Drives Productivity: Efficient Dataset Pruning for Enhanced Transfer Learningcited by 11
  2. 2023Adaptive Training Distributions with Scalable Online Bilevel Optimizationcited by 11
  3. 2024Source Matters: Source Dataset Impact on Model Robustness in Medical Imagingcited by 10
  4. 2024Label Smoothing Improves Machine Unlearningcited by 8
  5. 2024Transfer Learning Beyond Bounded Density Ratioscited by 6
  6. 2025Bridging Streaming Continual Learning via In-Context Large Tabular Modelscited by 6
  7. 2023Generating Synthetic Datasets by Interpolating along Generalized Geodesicscited by 5
  8. 2022Striving for data-model efficiency: Identifying data externalities on group performancecited by 4
  9. 2023Improving Few-shot Generalization of Safety Classifiers via Data Augmented Parameter-Efficient Fine-Tuningcited by 4
  10. 2024Conditional Language Learning with Contextcited by 4
  11. 2025In-Context Adaptation to Concept Drift for Learned Database Operationscited by 3
  12. 2024Unlearning Information Bottleneck: Machine Unlearning of Systematic Patterns and Biasescited by 2
  13. 2025Constraining Sequential Model Editing with Editing Anchor Compressioncited by 2
  14. 2025Image Coding for Machines via Feature-Preserving Rate-Distortion Optimizationcited by 1
  15. 2025Small-to-Large Generalization: Data Influences Models Consistently Across Scalecited by 1
  16. 2025Invisible Watermarks, Visible Gains: Steering Machine Unlearning with Bi-Level Watermarking Designcited by 1
  17. 2026Unified Optimization of Source Weights and Transfer Quantities in Multi-Source Transfer Learning: An Asymptotic Frameworkcited by 1
  18. 2025Escaping The Big Data Paradigm in Self-Supervised Representation Learningcited by 0
  19. 2025WSS-CL: Weight Saliency Soft-Guided Contrastive Learning for Efficient Machine Unlearning Image Classificationcited by 0
  20. 2025Intuitions of Machine Learning Researchers about Transfer Learning for Medical Image Classificationcited by 0