transferability

Transferable Decoding with Visual Entities for Zero-Shot Image Captioning, ICCV 2023

Efficiently find the best-suited language model (LM) for your NLP task

A Systematic Investigation of Transferability and Robustness of Humor Detection Models

Ensemble Adversarial Black-Box Attacks against Deep Learning Systems Trained by MNIST, USPS and GTSRB Datasets

SaTML 2023, 1st place in CVPR’21 Security AI Challenger: Unrestricted Adversarial Attacks on ImageNet.

The extension of "Patch-wise Attack for Fooling Deep Neural Network (ECCV2020)", and we aim to boost the success rates of targeted attack.

Our Team (green hand) 6th Solution for CVPR-2021 AIC-VI: Unrestricted Adversarial Attacks on ImageNet

Our simple but effective staircase sign method which boosts the transferability of both non-targeted and targeted attacks.

(AAAI 2024) Transferable Adversarial Attacks for Object Detection using Object-Aware Significant Feature Distortion

Source of the ECCV22 paper "LGV: Boosting Adversarial Example Transferability from Large Geometric Vicinity"

A Transferability-guided Protein-Ligand Interaction Prediction Method

[ICCV2023] ETran: Energy-based Transferability Estimation

[BMVC 2023] Diversifying the High-level Features for better Adversarial Transferability

Source of the UAI2022 paper "Efficient and Transferable Adversarial Examples from Bayesian Neural Networks"

Multi-source Transfer Learning method for time series with Shapelet Similarity-based Source Selection implemented with Keras

Reproducible CIFAR-10 study of adversarial transfer directionality, attack conditioning, and denominator-aware evaluation across ResNet-18, VGG16, and MobileNetV2.

Homework of Security and Privacy of Machine Learning (SPML Lectured by Shang-Tse Chen at NTU)

[ECCV 2026] Proteus: Model Leakage-Induced Adversarial Attack in Federated Learning

Matched-exposure benchmark for cross-event coseismic landslide segmentation (GNHR submission): code, frozen manifests, machine-readable metrics, audits. 同震滑坡跨事件迁移的匹配曝光基准可复现性包：四骨干对照、冻结清单与机器可读指标。

Reproducible PyTorch benchmark of FGSM and PGD attacks on GTSRB traffic-sign classifiers, with cross-model transfer evaluation, corner-PGD diagnostics, and documented results on Apple silicon.