Hang Su

53

Papers

63

Total Citations

Papers (53)

Exploring the Transferability of Visual Prompting for Multimodal Large Language Models

Controllable Navigation Instruction Generation with Chain of Thought Prompting

Zero-Shot Monocular Scene Flow Estimation in the Wild

Visual Generation Without Guidance

An N-Point Linear Solver for Line and Motion Estimation with Event Cameras

DPOT: Auto-Regressive Denoising Operator Transformer for Large-Scale PDE Pre-Training

Machine Vision Therapy: Multimodal Large Language Models Can Enhance Visual Robustness via Denoising In-Context Learning

Layer-Aware Analysis of Catastrophic Overfitting: Revealing the Pseudo-Robust Shortcut Dependency

Fourier Controller Networks for Real-Time Decision-Making in Embodied Learning

Improved Operator Learning by Orthogonal Attention

Reference Neural Operators: Learning the Smooth Dependence of Solutions of PDEs on Geometric Deformations

Active Sample Selection and Correction Propagation on a Gradually-Augmented Graph

Improving Interpretability of Deep Neural Networks With Semantic Information

SPLATNet: Sparse Lattice Networks for Point Cloud Processing

Textbook Question Answering Under Instructor Guidance With Memory Networks

Interpret Neural Networks by Identifying Critical Data Routing Paths

Boosting Adversarial Attacks With Momentum

Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attacks

Efficient Decision-Based Black-Box Adversarial Attacks on Face Recognition

Pixel-Adaptive Convolutional Neural Networks

Benchmarking Adversarial Robustness on Image Classification

LiBRe: A Practical Bayesian Approach to Adversarial Detection

QAIR: Practical Query-Efficient Black-Box Attacks for Image Retrieval

Unsupervised Part Segmentation Through Disentangling Appearance and Shape

Two Coupled Rejection Metrics Can Tell Adversarial Examples Apart

Towards Effective Adversarial Textured 3D Meshes on Physical Face Recognition

All Are Worth Words: A ViT Backbone for Diffusion Models

Benchmarking Robustness of 3D Object Detection to Common Corruptions

End-To-End Face Detection and Cast Grouping in Movies Using Erdos-Renyi Clustering

Black-Box Detection of Backdoor Attacks With Limited Information and Data

Towards Face Encryption by Generating Adversarial Identity Masks

A 5-Point Minimal Solver for Event Camera Relative Motion Estimation

Towards Viewpoint-Invariant Visual Recognition via Adversarial Training

COCO-O: A Benchmark for Object Detectors under Natural Distribution Shifts

Detection Transformer with Stable Matching

Training Interpretable Convolutional Neural Networks by Differentiating Class-specific Filters

Defense Against Adversarial Attacks via Controlling Gradient Leaking on Embedded Manifolds

Boosting Transferability of Targeted Adversarial Examples via Hierarchical Generative Networks

Multi-View Convolutional Neural Networks for 3D Shape Recognition

A Linear N-Point Solver for Structure and Motion from Asynchronous Tracks

AdvDreamer Unveils: Are Vision-Language Models Truly Ready for Real-World 3D Variations?

Towards Transferable Targeted 3D Adversarial Attack in the Physical World

Robust Classification via a Single Diffusion Model

Improving Black-box Adversarial Attacks with a Transfer-based Prior

Boosting Adversarial Training with Hypersphere Embedding

Adversarial Distributional Training for Robust Deep Learning

Bi-level Score Matching for Learning Energy-based Latent Variable Models

Accumulative Poisoning Attacks on Real-time Data

A Unified Hard-Constraint Framework for Solving Geometrically Complex PDEs

ViewFool: Evaluating the Robustness of Visual Recognition to Adversarial Viewpoints

ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation

Overcoming Recency Bias of Normalization Statistics in Continual Learning: Balance and Adaptation

Hierarchical Decomposition of Prompt-Based Continual Learning: Rethinking Obscured Sub-optimality