Minjing Dong

11

Papers

46

Total Citations

Papers (11)

VSSD: Vision Mamba with Non-Causal State Space Duality

Harnessing Vision Foundation Models for High-Performance, Training-Free Open Vocabulary Segmentation

Feature Clipping for Uncertainty Calibration

Uncertainty Weighted Gradients for Model Calibration

Beyond One-Hot Labels: Semantic Mixing for Model Calibration

Efficient Image-to-Image Diffusion Classifier for Adversarial Robustness

AMD: Autoregressive Motion Diffusion

Adversarial Robustness via Deformable Convolution with Stochasticity

Imitation Learning from Purified Demonstrations

Random Entangled Tokens for Adversarially Robust Vision Transformer

Backdooring Self-Supervised Contrastive Learning by Noisy Alignment