# Rethinking Model Ensemble in Transfer-based Adversarial Attacks

arXiv: https://arxiv.org/abs/2303.09105 · HTML: https://mlsanity.com/paper/2303.09105/

## Key facts

- Cited by 89 papers in the map (data through 14 Oct 2026).
- Builds on 72 of its 90 references that are in the map.
- First posted on arXiv 16 Mar 2023.

## Builds on (most cited first, 30 of 72)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [SSD: Single Shot MultiBox Detector](https://mlsanity.com/paper/1512.02325/) (2015)
- [Towards Evaluating the Robustness of Neural Networks](https://mlsanity.com/paper/1608.04644/) (2016)
- [YOLOv3: An Incremental Improvement](https://mlsanity.com/paper/1804.02767/) (2018)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [Adversarial examples in the physical world](https://mlsanity.com/paper/1607.02533/) (2016)
- [YOLOv4: Optimal Speed and Accuracy of Object Detection](https://mlsanity.com/paper/2004.10934/) (2020)
- [YOLO9000: Better, Faster, Stronger](https://mlsanity.com/paper/1612.08242/) (2016)
- [Sharpness-Aware Minimization for Efficiently Improving Generalization](https://mlsanity.com/paper/2010.01412/) (2020)
- [Certified Adversarial Robustness via Randomized Smoothing](https://mlsanity.com/paper/1902.02918/) (2019)
- [Averaging Weights Leads to Wider Optima and Better Generalization](https://mlsanity.com/paper/1803.05407/) (2018)
- [ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design](https://mlsanity.com/paper/1807.11164/) (2018)
- [Visualizing the Loss Landscape of Neural Nets](https://mlsanity.com/paper/1712.09913/) (2017)
- [SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size](https://mlsanity.com/paper/1602.07360/) (2016)

## Built on by (most cited first, 30 of 89)

- [How Robust is Google's Bard to Adversarial Image Attacks?](https://mlsanity.com/paper/2309.11751/) (2023)
- [Dissecting Adversarial Robustness of Multimodal LM Agents](https://mlsanity.com/paper/2406.12814/) (2024)
- [FlipAttack: Jailbreak LLMs via Flipping](https://mlsanity.com/paper/2410.02832/) (2024)
- [Robust Classification via a Single Diffusion Model](https://mlsanity.com/paper/2305.15241/) (2023)
- [JailbreakZoo: Survey, Landscapes, and Horizons in Jailbreaking Large Language and Vision-Language Models](https://mlsanity.com/paper/2407.01599/) (2024)
- [Manipulating Multimodal Agents via Cross-Modal Prompt Injection](https://mlsanity.com/paper/2504.14348/) (2025)
- [Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment](https://mlsanity.com/paper/2505.21494/) (2025)
- [Elucidating the Design Space of Dataset Condensation](https://mlsanity.com/paper/2404.13733/) (2024)
- [Boosting Jailbreak Attack with Momentum](https://mlsanity.com/paper/2405.01229/) (2024)
- [Chain of Attack: On the Robustness of Vision-Language Models Against Transfer-Based Adversarial Attacks](https://mlsanity.com/paper/2411.15720/) (2024)
- [Efficient Generation of Targeted and Transferable Adversarial Examples for Vision-Language Models Via Diffusion Models](https://mlsanity.com/paper/2404.10335/) (2024)
- [A Frustratingly Simple Yet Highly Effective Attack Baseline: Over 90% Success Rate Against the Strong Black-box Models of GPT-4.5/4o/o1](https://mlsanity.com/paper/2503.10635/) (2025)
- [Why Does Little Robustness Help? A Further Step Towards Understanding Adversarial Transferability](https://mlsanity.com/paper/2307.07873/) (2023)
- [A Transfer Attack to Image Watermarks](https://mlsanity.com/paper/2403.15365/) (2024)
- [Your Diffusion Model is Secretly a Certifiably Robust Classifier](https://mlsanity.com/paper/2402.02316/) (2024)
- [On the Duality Between Sharpness-Aware Minimization and Adversarial Training](https://mlsanity.com/paper/2402.15152/) (2024)
- [Adversarial Attacks of Vision Tasks in the Past 10 Years: A Survey](https://mlsanity.com/paper/2410.23687/) (2024)
- [ADVEDM:Fine-grained Adversarial Attack against VLM-based Embodied Agents](https://mlsanity.com/paper/2509.16645/) (2025)
- [X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP](https://mlsanity.com/paper/2505.05528/) (2025)
- [MirrorCheck: Efficient Adversarial Defense for Vision-Language Models](https://mlsanity.com/paper/2406.09250/) (2024)
- [Exploring the Transferability of Visual Prompting for Multimodal Large Language Models](https://mlsanity.com/paper/2404.11207/) (2024)
- [Sharpness-Aware Minimization Alone can Improve Adversarial Robustness](https://mlsanity.com/paper/2305.05392/) (2023)
- [Exploring the Robustness of In-Context Learning with Noisy Labels](https://mlsanity.com/paper/2404.18191/) (2024)
- [Prompt-Agnostic Adversarial Perturbation for Customized Diffusion Models](https://mlsanity.com/paper/2408.10571/) (2024)
- [Bag of Tricks to Boost Adversarial Transferability](https://mlsanity.com/paper/2401.08734/) (2024)
- [Adversarial Representation Engineering: A General Model Editing Framework for Large Language Models](https://mlsanity.com/paper/2404.13752/) (2024)
- [DAT: Improving Adversarial Robustness via Generative Amplitude Mix-up in Frequency Domain](https://mlsanity.com/paper/2410.12307/) (2024)
- [FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction](https://mlsanity.com/paper/2509.21029/) (2025)
- [Transferable Adversarial Attacks on Black-Box Vision-Language Models](https://mlsanity.com/paper/2505.01050/) (2025)
- [JailBound: Jailbreaking Internal Safety Boundaries of Vision-Language Models](https://mlsanity.com/paper/2505.19610/) (2025)

## Abstract (arXiv)

It is widely recognized that deep learning models lack robustness to adversarial examples. An intriguing property of adversarial examples is that they can transfer across different models, which enables black-box attacks without any knowledge of the victim model. An effective strategy to improve the transferability is attacking an ensemble of models. However, previous works simply average the outputs of different models, lacking an in-depth analysis on how and why model ensemble methods can strongly improve the transferability. In this paper, we rethink the ensemble in adversarial attacks and define the common weakness of model ensemble with two properties: 1) the flatness of loss landscape; and 2) the closeness to the local optimum of each model. We empirically and theoretically show that both properties are strongly correlated with the transferability and propose a Common Weakness Attack (CWA) to generate more transferable adversarial examples by promoting these two properties. Experimental results on both image classification and object detection tasks validate the effectiveness of our approach to improving the adversarial transferability, especially when attacking adversarially trained models. We also successfully apply our method to attack a black-box large vision-language model -- Google's Bard, showing the practical effectiveness. Code is available at urlhttps://github.com/huanranchen/AdversarialAttacks.