mlsanityData to 7 Oct 2026

Active Finetuning: Exploiting Annotation Budget in the Pretraining-Finetuning Paradigm

arXiv 2303.14382 ↗First posted 25 Mar 2023cs.CV cs.LGcited by 36

Not on the map. Why?

Abstract (arXiv)

Given the large-scale data and the high annotation cost, pretraining-finetuning becomes a popular paradigm in multiple computer vision tasks. Previous research has covered both the unsupervised pretraining and supervised finetuning in this paradigm, while little attention is paid to exploiting the annotation budget for finetuning. To fill in this gap, we formally define this new active finetuning task focusing on the selection of samples for annotation in the pretraining-finetuning paradigm. We propose a novel method called ActiveFT for active finetuning task to select a subset of data distributing similarly with the entire unlabeled pool and maintaining enough diversity by optimizing a parametric model in the continuous space. We prove that the Earth Mover's distance between the distributions of the selected subset and the entire data pool is also reduced in this process. Extensive experiments show the leading performance and high efficiency of ActiveFT superior to baselines on both image classification and semantic segmentation. Our code is released at https://github.com/yichen928/ActiveFT.

Key facts. Cited by 36 papers in the map (data through 7 Oct 2026). Builds on 40 of its 56 references that are in the map. First posted on arXiv 25 Mar 2023.

Builds on

40 of 56 references.

References · 40

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  5. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  6. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  7. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  8. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  9. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  10. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  11. 2020Training data-efficient image transformers & distillation through attentioncited by 4,560
  12. 2020Exploring Simple Siamese Representation Learningcited by 2,853
28 more
  1. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,845
  2. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,360
  3. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,222
  4. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,838
  5. 2017Decoupled Weight Decay Regularizationcited by 1,722
  6. 2021An Empirical Study of Training Self-Supervised Vision Transformerscited by 1,610
  7. 2021iBOT: Image BERT Pre-Training with Online Tokenizercited by 891
  8. 2021Segmenter: Transformer for Semantic Segmentationcited by 855
  9. 2021Masked Feature Prediction for Self-Supervised Visual Pre-Trainingcited by 573
  10. 2020A Survey of Deep Active Learningcited by 547
  11. 2019Learning Loss for Active Learningcited by 335
  12. 2019Variational Adversarial Active Learningcited by 284
  13. 2021Spatio-temporal Self-Supervised Representation Learning for 3D Point Cloudscited by 180
  14. 2017Cost-Effective Active Learning for Deep Image Classificationcited by 165
  15. 2020Contextual Diversity for Active Learningcited by 139
  16. 2022Contrastive Masked Autoencoders are Stronger Vision Learnerscited by 133
  17. 2020Task-Aware Variational Adversarial Active Learningcited by 84
  18. 2021Influence Selection for Active Learningcited by 82
  19. 2017Learning how to Active Learn: A Deep Reinforcement Learning Approachcited by 79
  20. 2021Multiple instance active learning for object detectioncited by 77
  21. 2018Efficient Active Learning for Image Classification and Segmentation using a Sample Selection and Conditional Generative Adversarial Networkcited by 74
  22. 2017Generative Adversarial Active Learningcited by 69
  23. 2022Active Learning by Feature Mixingcited by 63
  24. 2018Adversarial Sampling for Active Learningcited by 51
  25. 2021Semi-Supervised Active Learning with Temporal Output Discrepancycited by 50
  26. 2021Reducing Label Effort: Self-Supervised meets Active Learningcited by 42
  27. 2021Low Budget Active Learning via Wasserstein Distance: An Integer Programming Approachcited by 34
  28. 2022PT4AL: Using Self-Supervised Pretext Tasks for Active Learningcited by 27

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 13; 2025: 9; 2026: 7.

36 papers.

Sort:

By area:

  1. 2023A comprehensive survey on deep active learning in medical image analysiscited by 32
  2. 2024Multi-Task Model Merging via Adaptive Weight Disentanglementcited by 22
  3. 2025Reinforced Refinement with Self-Aware Expansion for End-to-End Autonomous Drivingcited by 19
  4. 2024ActiveAD: Planning-Oriented Active Learning for End-to-End Autonomous Drivingcited by 16
  5. 2024Generalized Coverage for More Robust Low-Budget Active Learningcited by 15
  6. 2024Active Learning for Vision-Language Modelscited by 15
  7. 2024Post-hoc Probabilistic Vision-Language Modelscited by 13
  8. 2024Fairness Without Harm: An Influence-Guided Active Sampling Approachcited by 12
  9. 2024Uncertainty Herding: One Active Learning Method for All Label Budgetscited by 12
  10. 2025Filter Images First, Generate Instructions Later: Pre-Instruction Data Selection for Visual Instruction Tuningcited by 12
  11. 2023Selectivity Drives Productivity: Efficient Dataset Pruning for Enhanced Transfer Learningcited by 11
  12. 2023Towards Free Data Selection with General-Purpose Modelscited by 10
24 more
  1. 2023BAL: Balancing Diversity and Novelty for Active Learningcited by 9
  2. 2025Rethinking Epistemic and Aleatoric Uncertainty for Active Open-Set Annotation: An Energy-Based Approachcited by 5
  3. 2024Feature Alignment: Rethinking Efficient Active Learning via Proxy in the Context of Pre-trained Modelscited by 4
  4. 2023NTKCPL: Active Learning on Top of Self-Supervised Model by Estimating True Coveragecited by 3
  5. 2023ActiveDC: Distribution Calibration for Active Finetuningcited by 3
  6. 2025Certainty and Uncertainty Guided Active Domain Adaptationcited by 3
  7. 2024VeCAF: Vision-language Collaborative Active Finetuning with Training Objective Awarenesscited by 2
  8. 2024DCoM: Active Learning for All Learnerscited by 2
  9. 2024Enhancing Semi-Supervised Learning via Representative and Diverse Sample Selectioncited by 2
  10. 2026VIOLA: Towards Video In-Context Learning with Minimal Annotationscited by 2
  11. 2024Enhancing Modality Representation and Alignment for Multimodal Cold-start Active Learningcited by 1
  12. 2025Efficient Code LLM Training via Distribution-Consistent and Diversity-Aware Data Selectioncited by 1
  13. 2026Labeled TrustSet Guided: Batch Active Learning with Reinforcement Learningcited by 1
  14. 2023Active Vision Reinforcement Learning under Limited Visual Observabilitycited by 0
  15. 2024Towards Clinical Practice in CT-Based Pulmonary Disease Screening: An Efficient and Reliable Frameworkcited by 0
  16. 2025Active Learning from Scene Embeddings for End-to-End Autonomous Drivingcited by 0
  17. 2025Optimizing Active Learning in Vision-Language Models via Parameter-Efficient Uncertainty Calibrationcited by 0
  18. 2025Combining Discrepancy-Confusion Uncertainty and Calibration Diversity for Active Fine-Grained Image Classificationcited by 0
  19. 2025Hierarchical Dataset Selection for High-Quality Data Sharingcited by 0
  20. 2026Revisiting Unknowns: Towards Effective and Efficient Open-Set Active Learningcited by 0
  21. 2026Boundary-Centric Clip-Budgeted Active Learning for Temporal Action Segmentationcited by 0
  22. 2026FACT: A Simple and Efficient Framework for Active Finetuningcited by 0
  23. 2026One Knob to Rule Them All: A Unified Optimal Transport View of Cold-Start Active Learningcited by 0
  24. 2026Active Learning for Biodiversity Monitoring: From Label Efficiency to Reliable Ecological Inferencecited by 0