Sotabase

Career

· Senior Research Scientist, Google2023–
· Postdoctoral Researcher, University of Oxford2023–

Publications (77)

A Systematic Survey of Prompt Engineering on Vision-Language Foundation Models
arXiv.org · 2023
216
cited
MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models
European Conference on Computer Vision · 2023
194
cited
Understanding Individual Decisions of CNNs via Contrastive Backpropagation
Asian Conference on Computer Vision · 2018
104
cited
SegPGD: An Effective and Efficient Adversarial Attack for Evaluating and Boosting Segmentation Robustness
European Conference on Computer Vision · 2022
92
cited
FedDAT: An Approach for Foundation Model Finetuning in Multi-Modal Heterogeneous Federated Learning
AAAI Conference on Artificial Intelligence · 2023
86
cited
Are Vision Transformers Robust to Patch Perturbations?
European Conference on Computer Vision · 2021
81
cited
Improved Techniques for Optimization-Based Jailbreaking on Large Language Models
International Conference on Learning Representations · 2024
80
cited
Backdoor Defense via Adaptively Splitting Poisoned Dataset
Computer Vision and Pattern Recognition · 2023
71
cited
Self-Discovering Interpretable Diffusion Latent Directions for Responsible Text-to-Image Generation
Computer Vision and Pattern Recognition · 2023
64
cited
Improving the Robustness of Capsule Networks to Image Affine Transformations
Computer Vision and Pattern Recognition · 2019
58
cited
Latent Guard: a Safety Framework for Text-to-image Generation
European Conference on Computer Vision · 2024
56
cited
A Survey on Transferability of Adversarial Examples across Deep Neural Networks
Trans. Mach. Learn. Res. · 2023
52
cited
Safety at Scale: A Comprehensive Survey of Large Model Safety
Robotics · 2025
48
cited
Towards Efficient Adversarial Training on Vision Transformers
European Conference on Computer Vision · 2022
45
cited
Interpretable Graph Capsule Networks for Object Recognition
AAAI Conference on Artificial Intelligence · 2020
41
cited
Attacking Adversarial Attacks as A Defense
arXiv.org · 2021
40
cited
Watermark Vaccine: Adversarial Attacks to Prevent Watermark Removal
European Conference on Computer Vision · 2022
34
cited
Capsule Network is Not More Robust than Convolutional Network
Computer Vision and Pattern Recognition · 2021
33
cited
Saliency Methods for Explaining Adversarial Attacks
arXiv.org · 2019
33
cited
Benchmarking Robustness of Adaptation Methods on Pre-trained Vision-Language Models
Neural Information Processing Systems · 2023
32
cited
Sotabase
Jindong Gu | Researcher Profile | Sotabase | Sotabase