mlsanityData to 14 Oct 2026

Model Leeching: An Extraction Attack Targeting LLMs

arXiv 2309.10544 ↗First posted 19 Sep 2023cs.LG cs.AI cs.CLcited by 22

Not on the map. Why?

Abstract (arXiv)

Model Leeching is a novel extraction attack targeting Large Language Models (LLMs), capable of distilling task-specific knowledge from a target LLM into a reduced parameter model. We demonstrate the effectiveness of our attack by extracting task capability from ChatGPT-3.5-Turbo, achieving 73% Exact Match (EM) similarity, and SQuAD EM and F1 accuracy scores of 75% and 87%, respectively for only 50 in API cost. We further demonstrate the feasibility of adversarial attack transferability from an extracted model extracted via Model Leeching to perform ML attack staging against a target LLM, resulting in an 11% increase to attack success rate when applied to ChatGPT-3.5-Turbo.

Key facts. Cited by 22 papers in the map (data through 14 Oct 2026). Builds on 23 of its 34 references that are in the map. First posted on arXiv 19 Sep 2023.

Builds on

23 of 34 references.

References · 23

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  4. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  5. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  6. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 3,104
  7. 2019ALBERT: A Lite BERT for Self-supervised Learning of Language Representationscited by 2,790
  8. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  9. 2023A Survey of Large Language Modelscited by 2,576
  10. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  11. 2020Extracting Training Data from Large Language Modelscited by 2,189
  12. 2015The Limitations of Deep Learning in Adversarial Settingscited by 790
11 more
  1. 2016Stealing Machine Learning Models via Prediction APIscited by 619
  2. 2023A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPTcited by 619
  3. 2017Adversarial Examples for Evaluating Reading Comprehension Systemscited by 571
  4. 2018Adversarial Attacks and Defences: A Surveycited by 259
  5. 2021Reframing Instructional Prompts to GPTk's Languagecited by 176
  6. 2019Thieves on Sesame Street! Model Extraction of BERT-based APIscited by 139
  7. 2019DAWN: Dynamic Adversarial Watermarking of Neural Networkscited by 100
  8. 2020The Turking Test: Can Language Models Understand Instructions?cited by 82
  9. 2022Machine Generated Text: A Comprehensive Survey of Threat Models and Detection Methodscited by 79
  10. 2022I Know What You Trained Last Summer: A Survey on Stealing Machine Learning Models and Defencescited by 75
  11. 2022PINCH: An Adversarial Extraction Attack Framework for Deep Learning Modelscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 8; 2025: 8; 2026: 5.

22 papers.

Sort:

By area:

  1. 2023A Survey of Text Watermarking in the Era of Large Language Modelscited by 112
  2. 2024Copyright Protection in Generative AI: A Technical Perspectivecited by 38
  3. 2024Securing Large Language Models: Threats, Vulnerabilities and Responsible Practicescited by 37
  4. 2025A Survey on Model Extraction Attacks and Defenses for Large Language Modelscited by 19
  5. 2024SLIP: Securing LLMs IP Using Weights Decompositioncited by 14
  6. 2025Attack and defense techniques in large language models: A survey and new perspectivescited by 14
  7. 2024Towards More Realistic Extraction Attacks: An Adversarial Perspectivecited by 8
  8. 2025LLM Security: Vulnerabilities, Attacks, Defenses, and Countermeasurescited by 8
  9. 2024MEFT: Memory-Efficient Fine-Tuning through Sparse Adaptercited by 5
  10. 2024Can LLMs be Fooled? Investigating Vulnerabilities in LLMscited by 5
  11. 2025Towards Harnessing the Collaborative Power of Large and Small Models for Domain Taskscited by 4
  12. 2025Black-Box Behavioral Distillation Breaks Safety Alignment in Medical LLMscited by 4
10 more
  1. 2024A Survey on Privacy Attacks Against Digital Twin Systems in AI-Roboticscited by 3
  2. 2026An Embarrassingly Simple Detector for Model Extraction Attacks in Large Language Model API Trafficcited by 3
  3. 2026Let Them Steal: Trapping Large Language Model Extraction Attacks with Knowledge Honeypotcited by 3
  4. 2025Securing Educational LLMs: A Generalised Taxonomy of Attacks on LLMs and DREAD Risk Assessmentcited by 2
  5. 2026LLM-enabled Applications Require System-Level Threat Monitoringcited by 2
  6. 2024Inner-Probe: Discovering Copyright-related Data Generation in LLM Architecturecited by 1
  7. 2025LLM in the Middle: A Systematic Review of Threats and Mitigations to Real-World LLM-based Systemscited by 1
  8. 2026Robust LLM Watermarking with Minimal Semantic Distortion for IP Protectioncited by 1
  9. 2025DERMARK: A Dynamic, Efficient and Robust Multi-bit Watermark for Large Language Modelscited by 0
  10. 2026JudgeStealer: Extracting LLM Judging Capabilities across Evaluation Protocolscited by 0