Browse State-of-the-Art › Model extraction
Model extraction
57 papers with code · 1 benchmark · 2 datasets archive 2025-07-28
Model extraction attacks, aka model stealing attacks, are used to extract the parameters from the target model. Ideally, the adversary will be able to steal and replicate a model that will have a very similar performance to the target model.
Description from the archive archive 2025-07-28.
Benchmarks archive 2025-07-28
1 leaderboard table shown for this task, 1 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted.
| Dataset | Best model (first row in archive order) | Paper | Code | Syntology | Compare |
|---|---|---|---|---|---|
| UML Classes With Specs (1 row) | three-step-original | Towards Automatically Extracting UML Class Diagrams from Natural... | code | — | Compare |
Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.
Libraries
Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.
Datasets archive 2025-07-28
2 datasets whose archive record lists this task, ordered by the archive's paper count.
Subtasks archive 2025-07-28
No subtask under this task in the archive's task tree.
Papers archive 2025-07-28
archive papers tagged: 176 · with a code link: 57 · where Syntology ran a sample: 12 (10 with a run with no instrument failure, 2 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (12 of 176 tagged: 10 with a run with no instrument failure, 2 where every run was a failure of Syntology's instrument)
Page 1 of 2: papers 1 to 100 of 176, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.
Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.
-
3 Dec 2022 3 repositories listed Syntology official (archive's flag): 2 ran · 7 ran (of which 2 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 8 unverified (of 15 harvested samples)Most cross-device federated learning (FL) studies focus on the model-homogeneous setting where the global server model and local client models are identical.
-
27 Feb 2020 3 repositories listedSuch pairs are watermarks, which are not sampled from the task distribution and are only known to the defender.
-
4 Sep 2024 2 repositories listed Syntology official (archive's flag): 6 ran · 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified (of 7 harvested samples) · 7 pointer-only (licence)Model extraction attacks (MEAs) on large language models (LLMs) have received increasing attention in recent research.
-
6 Feb 2023 2 repositories listedWe can then detect the secret message by probing a suspect model to tell if it is distilled from the protected one.
-
7 Oct 2021 2 repositories listedThe extraction of process models from text refers to the problem of turning the information contained in an unstructured textual process descriptions into a formal representation, i. e., a process model.
-
30 Nov 2020 2 repositories listedCurrent model extraction attacks assume that the adversary has access to a surrogate dataset with characteristics similar to the proprietary data used to train the victim model.
-
25 Oct 2020 2 repositories listedDeep Neural Networks (DNNs) have achieved remarkable performance on a range of tasks.
-
21 Jun 2025 1 repository listed Syntology official (archive's flag): 8 ran · 8 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 3 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified (of 10 harvested samples)Graph Neural Networks (GNNs) have demonstrated remarkable utility across diverse applications, and their growing complexity has made Machine Learning as a Service (MLaaS) a viable platform for scalable deployment.
-
3 Jun 2025 1 repository listedTo address this gap, we propose MISLEADER (enseMbles of dIStiLled modEls Against moDel ExtRaction), a novel defense strategy that does not rely on OOD assumptions.
-
20 Mar 2025 1 repository listed Syntology official (archive's flag): 3 ran · 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified (of 3 harvested samples)Graph Neural Networks (GNNs) have gained traction in Graph-based Machine Learning as a Service (GMLaaS) platforms, yet they remain vulnerable to graph-based model extraction attacks (MEAs), where adversaries reconstruct…
-
7 Feb 2025 1 repository listedThe advent of Machine Learning as a Service (MLaaS) has heightened the trade-off between model explainability and security.
-
2 Feb 2025 1 repository listedThe rapid advancement of large models, driven by their exceptional abilities in learning and generalization through large-scale pre-training, has reshaped the landscape of Artificial Intelligence (AI).
-
16 Jan 2025 1 repository listedGuided by the model, we further introduce: (1) a similarity-based training-free watermarking method for plug-and-play and flexible watermarking, and (2) a distribution-based multi-step watermark information transmission…
-
15 Nov 2024 1 repository listedIn this work, we introduce a new end-to-end attack framework designed for model extraction of embedded DNNs with high fidelity.
-
14 Nov 2024 1 repository listedEmbedding-as-a-Service (EaaS) has emerged as a successful business pattern but faces significant challenges related to various forms of copyright infringement, particularly, the API misuse and model extraction attacks.
-
23 Oct 2024 1 repository listedMotivated by this, in this paper, we propose a novel embedding-specific watermarking (ESpeW) mechanism to offer robust copyright protection for EaaS.
-
21 Sep 2024 1 repository listedModel extraction aims to create a functionally similar copy from a machine learning as a service (MLaaS) API with minimal overhead, typically for illicit profit or as a precursor to further attacks, posing a significant…
-
4 Aug 2024 1 repository listedIn the domain of black-box model extraction, conventional methods reliant on soft labels or surrogate datasets struggle with scaling to high-dimensional input spaces and managing the complexity of an extensive array of…
-
14 Jun 2024 1 repository listedOur study evaluates the feasibility of parameter extraction methods of Carlini et al. [1] further enhanced by Canales-Martínez et al. [2] for models trained on standard benchmarks.
-
29 May 2024 1 repository listedCounterfactual (CF) explanations for ML model predictions provide actionable recourse recommendations to individuals adversely impacted by predicted outcomes.
-
8 May 2024 1 repository listed Syntology official: harvested, nothing ran · 0 ran · 1 unverified (of 1 harvested sample)However, counterfactual explanations can also be leveraged to reconstruct the model by strategically training a surrogate model to give similar predictions as the original (target) model.
-
4 Apr 2024 1 repository listedXAI techniques aim to enhance the transparency of ML models by providing insights, in terms of model's explanations, into their decision-making process.
-
15 Mar 2024 1 repository listedUnfortunately, existing methodologies based on trigger sets are still susceptible to functionality-stealing attacks, potentially enabling adversaries to steal the functionality of the source model without a reliable…
-
3 Mar 2024 1 repository listedEmbedding as a Service (EaaS) has become a widely adopted solution, which offers feature extraction capabilities for addressing various downstream tasks in Natural Language Processing (NLP).
-
26 Jan 2024 1 repository listedTo protect the Intellectual Property (IP) of the original owners over such DNN models, backdoor-based watermarks have been extensively studied.
-
17 Dec 2023 1 repository listedWhile deep learning models have shown significant performance across various domains, their deployment needs extensive resources and advanced computing infrastructure.
-
10 Nov 2023 1 repository listedOur extensive experiments on various datasets indicate that the proposed watermarking approach is effective and safe for verifying the copyright of VLPs for multi-modal EaaS and robust against model extraction attacks.
-
8 Nov 2023 1 repository listedIn this work, we explore the usage of an ensemble of deep learning models as our thief model.
-
25 Oct 2023 1 repository listedThese gradients are used to compute a swap loss, which maximizes the loss of the student model.
-
21 Oct 2023 1 repository listed Syntology official (archive's flag): 5 ran · 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified (of 7 harvested samples)We study model extraction attacks in natural language processing (NLP) where attackers aim to steal victim models by repeatedly querying the open Application Programming Interfaces (APIs).
-
4 Sep 2023 1 repository listed
-
31 Aug 2023 1 repository listed
-
19 Jul 2023 1 repository listed
-
5 Jul 2023 1 repository listed Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample)
-
Weighted Automata Extraction and Explanation of Recurrent Neural Networks for Natural Language Tasks24 Jun 2023 1 repository listed Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample) · 1 pointer-only (licence)
-
17 May 2023 1 repository listed Syntology official (archive's flag): 6 ran · 6 ran (of which 1 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified (of 10 harvested samples) · 1 pointer-only (licence)
-
16 Feb 2023 1 repository listed Syntology official (archive's flag): 5 ran · 5 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified (of 8 harvested samples)
-
26 Oct 2022 1 repository listed
-
16 May 2022 1 repository listed Syntology official: harvested, nothing ran · 0 ran · 2 unverified (of 2 harvested samples)
-
13 Apr 2022 1 repository listed
-
25 Feb 2022 1 repository listed
-
5 Dec 2021 1 repository listed
-
1 Sep 2021 1 repository listed
-
12 Jul 2021 1 repository listed
-
18 Mar 2021 1 repository listed
-
13 Dec 2020 1 repository listed Syntology official (archive's flag): 7 ran · 7 ran (of which 5 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 7 unverified (of 14 harvested samples) · 14 pointer-only (licence)
-
24 Oct 2020 1 repository listed
-
15 Oct 2020 1 repository listed
-
3 Sep 2020 1 repository listed
-
16 Apr 2020 1 repository listed
-
10 Mar 2020 1 repository listed Syntology official (archive's flag): 7 ran · 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified (of 9 harvested samples)
-
7 Feb 2020 1 repository listed
-
2 Dec 2019 1 repository listed
-
27 Oct 2019 1 repository listed
-
3 Jun 2019 1 repository listed Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample)
-
16 May 2019 1 repository listed
-
9 Sep 2016 1 repository listed
-
Entangled Threats: A Unified Kill Chain Model for Quantum Machine Learning Security11 Jul 2025 0 repositories listed
-
Navigating the Deep: Signature Extraction on Deep Neural Networks20 Jun 2025 0 repositories listed
-
GradEscape: A Gradient-Based Evader Against AI-Generated Text Detectors9 Jun 2025 0 repositories listed
-
Explore the vulnerability of black-box models via diffusion models9 Jun 2025 0 repositories listed
-
Evaluating Query Efficiency and Accuracy of Transfer Learning-based Model Extraction Attack in Federated Learning25 May 2025 0 repositories listed
-
On the interplay of Explainability, Privacy and Predictive Performance with Explanation-assisted Model Extraction13 May 2025 0 repositories listed
-
Better Decisions through the Right Causal World Model9 Apr 2025 0 repositories listed
-
CopyQNN: Quantum Neural Network Extraction Attack under Varying Quantum Noise1 Apr 2025 0 repositories listed
-
ProDiF: Protecting Domain-Invariant Features to Secure Pre-Trained Models Against Extraction17 Mar 2025 0 repositories listed
-
A Survey of Model Extraction Attacks and Defenses in Distributed Computing Environments22 Feb 2025 0 repositories listed
-
Differentially private fine-tuned NF-Net to predict GI cancer type17 Feb 2025 0 repositories listed
-
A Framework for Double-Blind Federated Adaptation of Foundation Models3 Feb 2025 0 repositories listed
-
Data-Free Model-Related Attacks: Unleashing the Potential of Generative AI28 Jan 2025 0 repositories listed
-
"FRAME: Forward Recursive Adaptive Model Extraction -- A Technique for Advance Feature Selection"21 Jan 2025 0 repositories listed
-
HoneypotNet: Backdoor Attacks Against Model Extraction2 Jan 2025 0 repositories listed
-
Bounding-box Watermarking: Defense against Model Extraction Attacks on Object Detectors20 Nov 2024 0 repositories listed
-
Few-shot Model Extraction Attacks against Sequential Recommender Systems18 Nov 2024 0 repositories listed
-
Efficient Model Extraction via Boundary Sampling20 Oct 2024 0 repositories listed
-
CaBaGe: Data-Free Model Extraction using ClAss BAlanced Generator Ensemble16 Sep 2024 0 repositories listed
-
Protecting Copyright of Medical Pre-trained Language Models: Training-Free Backdoor Model Watermarking14 Sep 2024 0 repositories listed
-
Enhancing TinyML Security: Study of Adversarial Attack Transferability16 Jul 2024 0 repositories listed
-
QUEEN: Query Unlearning against Model Extraction1 Jul 2024 0 repositories listed
-
Privacy Implications of Explainable AI in Data-Driven Systems22 Jun 2024 0 repositories listed
-
GENIE: Watermarking Graph Neural Networks for Link Prediction7 Jun 2024 0 repositories listed
-
Noisy Data Meets Privacy: Training Local Models with Post-Processed Remote Queries25 May 2024 0 repositories listed
-
DeepNcode: Encoding-Based Protection against Bit-Flip Attacks on Neural Networks22 May 2024 0 repositories listed
-
Learnable Linguistic Watermarks for Tracing Model Extraction Attacks on Large Language Models28 Apr 2024 0 repositories listed
-
QuantumLeak: Stealing Quantum Neural Networks from Cloud-based NISQ Machines16 Mar 2024 0 repositories listed
-
Precise Extraction of Deep Learning Models via Side-Channel Attacks on Edge/Endpoint Devices5 Mar 2024 0 repositories listed
-
Unraveling Attacks in Machine Learning-based IoT Ecosystems: A Survey and the Open Libraries Behind Them22 Jan 2024 0 repositories listed
-
MEAOD: Model Extraction Attack against Object Detectors22 Dec 2023 0 repositories listed
-
Model Extraction Attacks Revisited8 Dec 2023 0 repositories listed
-
Security and Privacy Challenges in Deep Learning Models23 Nov 2023 0 repositories listed
-
Like an Open Book? Read Neural Network Architecture with Simple Power Analysis on 32-bit Microcontrollers2 Nov 2023 0 repositories listed
-
Towards dialogue based, computer aided software requirements elicitation21 Oct 2023 0 repositories listed
-
SCME: A Self-Contrastive Method for Data-free and Query-Limited Model Extraction Attack15 Oct 2023 0 repositories listed
-
Beyond Labeling Oracles: What does it mean to steal ML models?3 Oct 2023 0 repositories listed
-
Towards Few-Call Model Stealing via Active Self-Paced Knowledge Distillation and Diffusion-Based Image Generation29 Sep 2023 0 repositories listed
-
Defending against Data-Free Model Extraction by Distributionally Robust Defensive Training21 Sep 2023 0 repositories listed
-
Defending against Data-Free Model Extraction by Distributionally Robust Defensive Training21 Sep 2023 0 repositories listed
-
Business Process Text Sketch Automation Generation Using Large Language Model3 Sep 2023 0 repositories listed
-
Fault Injection and Safe-Error Attack for Extraction of Embedded Neural Network Models31 Aug 2023 0 repositories listed
-
Data-Free Model Extraction Attacks in the Context of Object Detection9 Aug 2023 0 repositories listed
Syntology lines on 14 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections