Yahoo Suche Web Suche

Suchergebnisse

1. Suchergebnisse:

1. dl.acm.org › doi › pdfAdversarial Example Detection Using Robustness against Multi-Step...

   dl.acm.org › doi › pdf

   Vor 6 Tagen · Ian J. Goodfellow, Jonathon Shlens, and Christian Szegedy. 2015. Explaining and Harnessing Adversarial Examples. https://arxiv.org/abs/1412.6572. A Poster Paper at International Conference on Learning Representations.

2. link.springer.com › article › 10A Systematic Review on Generative Adversarial Network (GAN ... -...

   link.springer.com › article › 10

   • Im Cache

   14. Mai 2024 · A generative adversarial network was introduced in 2014 by researchers Ian J GoodFellow et al. in his research paper published in IEEE Journal. Most neural networks aim to learn from the limited data set, which usually faces misclassification and overfitting problems.

3. Bilder

   • 
   • 
   • 

   Alle anzeigen

4. arxiv.org › html › 2405Towards Transferable Attacks Against Vision-LLMs - arXiv.org

   arxiv.org › html › 2405

   • Im Cache

   23. Mai 2024 · Goodfellow et al. proposed the Fast Gradient Sign Method (FGSM ) to generate adversarial examples using a single gradient step, perturbing the model’s input before backpropagation.

5. dl.acm.org › doi › 10Physical Attack for Stereo Matching | Proceedings of the...

   dl.acm.org › doi › 10

   Vor 4 Tagen · So far, physical world attacks against deep stereo networks have not been systematically studied. Therefore, we extend the physical world attack to the domain of stereo matching. We design a new patch attack method for stereo matching, the disparity map attack.

6. arxiv.org › html › 2405Adversarial Robustness for Visual Grounding of Multimodal Large...

   arxiv.org › html › 2405

   • Im Cache

   16. Mai 2024 · The main contributions of this work are three-fold: (1) To the best of our knowledge, we are the first to reveal the adversarial threat in visual grounding of MLLMs.

7. dl.acm.org › doi › 10Adversarial Item Promotion on Visually-Aware Recommender Systems...

   dl.acm.org › doi › 10

   28. Mai 2024 · To shed light on the real vulnerabilities of visually-aware recommender systems when confronted with adversarial images, this paper introduces a novel attack method, IPDGI (Item Promotion by Diffusion Generated Image).

8. arxiv.org › html › 2405From Generalist to Specialist: Improving Large Language

   arxiv.org › html › 2405

   • Im Cache

   17. Mai 2024 · To address these concerns, four major solutions exist with varying degrees of complexity: (1) training a new language model from scratch, (2) fine-tuning pre-trained models, (3) Retrieval Augmented Generation (RAG) and (4) prompt engineering.