# Active Learning Helps Pretrained Models Learn the Intended Task

arXiv: https://arxiv.org/abs/2204.08491 · HTML: https://mlsanity.com/paper/2204.08491/

## Key facts

- Cited by 33 papers in the map (data through 14 Oct 2026).
- Builds on 44 of its 81 references that are in the map.
- First posted on arXiv 18 Apr 2022.

## Builds on (most cited first, 30 of 44)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization](https://mlsanity.com/paper/1911.08731/) (2019)
- [Deep Bayesian Active Learning with Image Data](https://mlsanity.com/paper/1703.02910/) (2017)
- [Big Transfer (BiT): General Visual Representation Learning](https://mlsanity.com/paper/1912.11370/) (2019)
- [Just Train Twice: Improving Group Robustness without Training Group Information](https://mlsanity.com/paper/2107.09044/) (2021)
- [Scaling Laws for Autoregressive Generative Modeling](https://mlsanity.com/paper/2010.14701/) (2020)
- [Underspecification Presents Challenges for Credibility in Modern Machine Learning](https://mlsanity.com/paper/2011.03395/) (2020)
- [Using Pre-Training Can Improve Model Robustness and Uncertainty](https://mlsanity.com/paper/1901.09960/) (2019)
- [True Few-Shot Learning with Language Models](https://mlsanity.com/paper/2105.11447/) (2021)
- [Functional Map of the World](https://mlsanity.com/paper/1711.07846/) (2017)
- [Calibration of Pre-trained Transformers](https://mlsanity.com/paper/2003.07892/) (2020)
- [An Investigation of Why Overparameterization Exacerbates Spurious Correlations](https://mlsanity.com/paper/2005.04345/) (2020)
- [Probabilistic Model-Agnostic Meta-Learning](https://mlsanity.com/paper/1806.02817/) (2018)
- [Simple data balancing achieves competitive worst-group-accuracy](https://mlsanity.com/paper/2110.14503/) (2021)
- [PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation](https://mlsanity.com/paper/2010.05903/) (2020)

## Built on by (most cited first, 30 of 33)

- [Selective Annotation Makes Language Models Better Few-Shot Learners](https://mlsanity.com/paper/2209.01975/) (2022)
- [Plex: Towards Reliability using Pretrained Large Model Extensions](https://mlsanity.com/paper/2207.07411/) (2022)
- [Eliciting Human Preferences with Language Models](https://mlsanity.com/paper/2310.11589/) (2023)
- [Active Learning Principles for In-Context Learning with Large Language Models](https://mlsanity.com/paper/2305.14264/) (2023)
- [A Survey of Active Learning for Natural Language Processing](https://mlsanity.com/paper/2210.10109/) (2022)
- [Task Ambiguity in Humans and Language Models](https://mlsanity.com/paper/2212.10711/) (2022)
- [What Do Learning Dynamics Reveal About Generalization in LLM Reasoning?](https://mlsanity.com/paper/2411.07681/) (2024)
- [ScatterShot: Interactive In-context Example Curation for Text Transformation](https://mlsanity.com/paper/2302.07346/) (2023)
- [LabelBench: A Comprehensive Framework for Benchmarking Adaptive Label-Efficient Learning](https://mlsanity.com/paper/2306.09910/) (2023)
- [Revisiting Active Learning in the Era of Vision Foundation Models](https://mlsanity.com/paper/2401.14555/) (2024)
- [AAPL: Adding Attributes to Prompt Learning for Vision-Language Models](https://mlsanity.com/paper/2404.16804/) (2024)
- [AmbigNLG: Addressing Task Ambiguity in Instruction for NLG](https://mlsanity.com/paper/2402.17717/) (2024)
- [Making Better Use of Unlabelled Data in Bayesian Active Learning](https://mlsanity.com/paper/2404.17249/) (2024)
- [Feature Dropout: Revisiting the Role of Augmentations in Contrastive Learning](https://mlsanity.com/paper/2212.08378/) (2022)
- [On the Limitations of Simulating Active Learning](https://mlsanity.com/paper/2305.13342/) (2023)
- [MultiBreak: A Scalable and Diverse Multi-turn Jailbreak Benchmark for Evaluating LLM Safety](https://mlsanity.com/paper/2605.01687/) (2026)
- [Active Test-time Vision-Language Navigation](https://mlsanity.com/paper/2506.06630/) (2025)
- [INTERACT: Enabling Interactive, Question-Driven Learning in Large Language Models](https://mlsanity.com/paper/2412.11388/) (2024)
- [Interactive Multi-fidelity Learning for Cost-effective Adaptation of Language Model with Sparse Human Supervision](https://mlsanity.com/paper/2310.20153/) (2023)
- [Active Learning for Multilingual Fingerspelling Corpora](https://mlsanity.com/paper/2309.12443/) (2023)
- [Active Few-Shot Fine-Tuning](https://mlsanity.com/paper/2402.15441/) (2024)
- [cRedAnno+: Annotation Exploitation in Self-Explanatory Lung Nodule Diagnosis](https://mlsanity.com/paper/2210.16097/) (2022)
- [Maximizing Information Gain in Privacy-Aware Active Learning of Email Anomalies](https://mlsanity.com/paper/2405.07440/) (2024)
- [No Free Lunch in Active Learning: LLM Embedding Quality Dictates Query Strategy Success](https://mlsanity.com/paper/2506.01992/) (2025)
- [Uncertainty estimation for out-of-distribution detection in computational histopathology](https://mlsanity.com/paper/2210.09909/) (2022)
- [Contextual Reliability: When Different Features Matter in Different Contexts](https://mlsanity.com/paper/2307.10026/) (2023)
- [Benchmarking Multi-Domain Active Learning on Image Classification](https://mlsanity.com/paper/2312.00364/) (2023)
- [Object-Centric Cropping for Visual Few-Shot Classification](https://mlsanity.com/paper/2508.00218/) (2025)
- [Decoupling Augmentation Bias in Prompt Learning for Vision-Language Models](https://mlsanity.com/paper/2511.03367/) (2025)
- [Socratic Students: Teaching Language Models to Learn by Asking Questions](https://mlsanity.com/paper/2512.13102/) (2025)

## Abstract (arXiv)

Models can fail in unpredictable ways during deployment due to task ambiguity, when multiple behaviors are consistent with the provided training data. An example is an object classifier trained on red squares and blue circles: when encountering blue squares, the intended behavior is undefined. We investigate whether pretrained models are better active learners, capable of disambiguating between the possible tasks a user may be trying to specify. Intriguingly, we find that better active learning is an emergent property of the pretraining process: pretrained models require up to 5 times fewer labels when using uncertainty-based active learning, while non-pretrained models see no or even negative benefit. We find these gains come from an ability to select examples with attributes that disambiguate the intended behavior, such as rare product categories or atypical backgrounds. These attributes are far more linearly separable in pretrained model's representation spaces vs non-pretrained models, suggesting a possible mechanism for this behavior.