mlsanityData to 14 Oct 2026

Revisiting Uncertainty-based Query Strategies for Active Learning with Transformers

arXiv 2107.05687 ↗First posted 12 Jul 2021cs.CL cs.LGcited by 60

Not on the map. Why?

Abstract (arXiv)

Active learning is the iterative construction of a classification model through targeted labeling, enabling significant labeling cost savings. As most research on active learning has been carried out before transformer-based language models ("transformers") became popular, despite its practical importance, comparably few papers have investigated how transformers can be combined with active learning to date. This can be attributed to the fact that using state-of-the-art query strategies for transformers induces a prohibitive runtime overhead, which effectively nullifies, or even outweighs the desired cost savings. For this reason, we revisit uncertainty-based query strategies, which had been largely outperformed before, but are particularly suited in the context of fine-tuning transformers. In an extensive evaluation, we connect transformers to experiments from previous research, assessing their performance on five widely used text classification benchmarks. For active learning with transformers, several other uncertainty-based approaches outperform the well-known prediction entropy query strategy, thereby challenging its status as most popular uncertainty baseline in active learning for text classification.

Key facts. Cited by 60 papers in the map (data through 14 Oct 2026). Builds on 17 of its 42 references that are in the map. First posted on arXiv 12 Jul 2021.

Builds on

17 of 42 references.

References · 17

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2017On Calibration of Modern Neural Networkscited by 4,960
  4. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,375
  5. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,307
  6. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  7. 2019XLNet: Generalized Autoregressive Pretraining for Language Understandingcited by 2,811
  8. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  9. 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,173
  10. 2014Convolutional Neural Networks for Sentence Classificationcited by 1,127
  11. 2019How to Fine-Tune BERT for Text Classification?cited by 411
  12. 2020Cold-start Active Learning through Self-supervised Language Modelingcited by 128
5 more
  1. 2018Practical Obstacles to Deploying Active Learningcited by 68
  2. 2019Sampling Bias in Deep Active Classification: An Empirical Studycited by 36
  3. 2015Self-Adaptive Hierarchical Sentence Modelcited by 10
  4. 2020Investigating the Effectiveness of Representations Based on Pretrained Transformer-based Language Models in Active Learning for Labelling Text Datasetscited by 6
  5. 2018Recurrently Controlled Recurrent Networkscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 3; 2022: 10; 2023: 21; 2024: 12; 2025: 12; 2026: 2.

60 papers.

Sort:

By area:

  1. 2023Active Prompting with Chain-of-Thought for Large Language Modelscited by 154
  2. 2024A Survey on Deep Active Learning: Recent Advances and New Frontierscited by 54
  3. 2021A Survey on Green Deep Learningcited by 52
  4. 2022A Survey of Active Learning for Natural Language Processingcited by 40
  5. 2023Which Examples to Annotate for In-Context Learning? Towards Effective and Efficient Selectioncited by 37
  6. 2023Large Language Models as Annotators: Enhancing Generalization of NLP Models at Minimal Costcited by 33
  7. 2022MEAL: Stable and Active Learning for Few-Shot Promptingcited by 20
  8. 2021Small-Text: Active Learning for Text Classification in Pythoncited by 19
  9. 2024ActiveLLM: Large Language Model-based Active Learning for Textual Few-Shot Scenarioscited by 18
  10. 2022Cold-Start Data Selection for Few-shot Language Model Fine-tuning: A Prompt-Based Uncertainty Propagation Approachcited by 16
  11. 2023ActiveGLAE: A Benchmark for Deep Active Learning with Transformerscited by 14
  12. 2023More Samples or More Prompts? Exploring Effective In-Context Sampling for LLM Few-Shot Prompt Engineeringcited by 13
38 more
  1. 2023Human Still Wins over LLM: An Empirical Study of Active Learning on Domain-Specific Annotation Taskscited by 13
  2. 2022Active Few-Shot Learning with FASLcited by 11
  3. 2023On the Limitations of Simulating Active Learningcited by 9
  4. 2023How Many Validation Labels Do You Need? Exploring the Design Space of Label-Efficient Model Rankingcited by 8
  5. 2024Not All Preference Pairs Are Created Equal: A Recipe for Annotation-Efficient Iterative Preference Learningcited by 7
  6. 2022The Infinite Index: Information Retrieval on Generative Text-To-Image Modelscited by 6
  7. 2023Investigating Multi-source Active Learning for Natural Language Inferencecited by 6
  8. 2022Active PETs: Active Data Annotation Prioritisation for Few-Shot Claim Verification with Pattern Exploiting Trainingcited by 5
  9. 2023Parameter-Efficient Language Model Tuning with Active Learning in Low-Resource Settingscited by 5
  10. 2023Learning to Learn for Few-shot Continual Active Learningcited by 5
  11. 2024Automatic Combination of Sample Selection Strategies for Few-Shot Learningcited by 5
  12. 2024STAR: Constraint LoRA with Dynamic Active Learning for Data-Efficient Fine-Tuning of Large Language Modelscited by 5
  13. 2024AnchorAL: Computationally Efficient Active Learning for Large and Imbalanced Datasetscited by 5
  14. 2022Is More Data Better? Re-thinking the Importance of Efficiency in Abusive Language Detection with Transformers-Based Active Learningcited by 4
  15. 2022Smooth Sailing: Improving Active Learning for Pre-trained Language Models with Representation Smoothness Analysiscited by 4
  16. 2023An Expanded Benchmark that Rediscovers and Affirms the Edge of Uncertainty Sampling for Active Learning in Tabular Datasetscited by 4
  17. 2023Cache & Distil: Optimising API Calls to Large Language Modelscited by 4
  18. 2023Interactive Multi-fidelity Learning for Cost-effective Adaptation of Language Model with Sparse Human Supervisioncited by 4
  19. 2022To Softmax, or not to Softmax: that is the question when applying Active Learning for Transformer Modelscited by 3
  20. 2023On Dataset Transferability in Active Learning for Transformerscited by 3
  21. 2023Active Learning for Natural Language Generationcited by 3
  22. 2023Divide & Conquer for Entailment-aware Multi-hop Evidence Retrievalcited by 3
  23. 2025DUAL: Diversity and Uncertainty Active Learning for Text Summarizationcited by 3
  24. 2024Active Learning for NLP with Large Language Modelscited by 2
  25. 2024On the Fragility of Active Learners for Text Classificationcited by 2
  26. 2024Active Learning for WBAN-based Health Monitoringcited by 2
  27. 2025DEUCE: Dual-diversity Enhancement and Uncertainty-awareness for Cold-start Active Learningcited by 2
  28. 2025Enhancing RAG with Active Learning on Conversation Records: Reject Incapables and Answer Capablescited by 2
  29. 2025Reassessing Active Learning Adoption in Contemporary NLP: A Community Surveycited by 2
  30. 2023XAL: EXplainable Active Learning Makes Classifiers Better Low-resource Learnerscited by 1
  31. 2024Label-Free Model Failure Detection for Lidar-based Point Cloud Segmentationcited by 1
  32. 2025Your Model is Overconfident, and Other Lies We Tell Ourselvescited by 1
  33. 2025No Free Lunch in Active Learning: LLM Embedding Quality Dictates Query Strategy Successcited by 1
  34. 2025InSQuAD: In-Context Learning for Efficient Retrieval via Submodular Mutual Information to Enforce Quality and Diversitycited by 1
  35. 2025Op-Fed: Opinion, Stance, and Monetary Policy Annotations on FOMC Transcripts Using Active Learningcited by 1
  36. 2021Active Learning for Argument Mining: A Practical Approachcited by 0
  37. 2022Active Learning with Tabular Language Modelscited by 0
  38. 2023Making Pre-trained Language Models both Task-solvers and Self-calibratorscited by 0

50 most cited of 60 All 60 (JSON)