CLIP-Guided Generative Networks for Transferable Targeted Adversarial Attacks.- Flash Cache: Reducing Bias in Radiance Cache Based Inverse Rendering.- Progressive Classifier and Feature Extractor Adaptation for Unsupervised Domain Adaptation on Point Clouds.- A New Dataset and Framework for Real-World Blurred Images Super-Resolution.- AddressCLIP: Empowering Vision-Language Models for City-wide Image Address Localization.- RISurConv: Rotation Invariant Surface Attention-Augmented Convolutions for 3D Point Cloud Classification and Segmentation.- StyleTokenizer: Defining Image Style by a Single Instance for Controlling Diffusion Models.- Bidirectional Uncertainty-Based Active Learning for Open-Set Annotation.
- Preventing Catastrophic Overfitting in Fast Adversarial Training: A Bi-level Optimization Perspective.- Projecting Points to Axes: Oriented Object Detection via Point-Axis Representation.- SeiT++: Masked Token Modeling Improves Storage-efficient Training.- Rectify the Regression Bias in Long-Tailed Object Detection.- MagicEraser: Erasing Any Objects via Semantics-Aware Control.- Reliable Spatial-Temporal Voxels For Multi-Modal Test-Time Adaptation.- Stable Preference: Redefining training paradigm of human preference model for Text-to-Image Synthesis.- SparseSSP: 3D Subcellular Structure Prediction from Sparse-View Transmitted Light Images.
- NL2Contact: Natural Language Guided 3D Hand-Object Contact Modeling with Diffusion Model.- Self-Adapting Large Visual-Language Models to Edge Devices across Visual Modalities.- Diff-Tracker: Text-to-Image Diffusion Models are Unsupervised Trackers.- Rethinking Tree-Ring Watermarking for Enhanced Multi-Key Identification.- 3D Small Object Detection with Dynamic Spatial Pruning.- STSP: Spatial-Temporal Subspace Projection for Video Class-incremental Learning.- Transferable 3D Adversarial Shape Completion using Diffusion Models.- OmniSat: Self-Supervised Modality Fusion for Earth Observation.
- Distilling Diffusion Models into Conditional GANs.- Semantically Guided Representation Learning For Action Anticipation.- MemBN: Robust Test-Time Adaptation via Batch Norm with Statistics Memory.