Yaowei Wang

Connecting Where You Look With What You Understand: Trajectory-Driven Localized Understanding for Interactive Vision-Language Models

VideoAnchor: Reinforcing Subspace-Structured Visual Cues for Coherent Visual-Spatial Reasoning

CoPRS: Learning Positional Prior from Chain-of-Thought for Reasoning Segmentation

CAI: Caption-Sensitive Attention Intervention for Mitigating Object Hallucination in Large Vision-Language Models

Norm$\times$Direction: Restoring the Missing Query Norm in Vision Linear Attention

Rectified Decoupled Dataset Distillation: A Closer Look for Fair and Comprehensive Evaluation

PIC: Revisiting INR for Image Coding with Fast Encoding and Sub-Millisecond Decoding

OVRD: Open-Vocabulary Relation DINO with Text-guided Salient Query Selection

In Defense of Prompt-based Continual Learning: Task Interference Mitigation via Confidence-Stratified Classifier Calibration

One Stone Three Birds: Training-free Core-context-aware Attention for Efficient LLM Prefilling, Decoding, and KV Caching

Spatial Understanding from Videos: Structured Prompts Meet Simulation Data

LoRATv2: Enabling Low-Cost Temporal Modeling in One-Stream Trackers

FOCUS: Unified Vision-Language Modeling for Interactive Editing Driven by Referential Segmentation

Perceptually Constrained Precipitation Nowcasting Model

Learning Spatial-Semantic Features for Robust Video Object Segmentation

An Exploration with Entropy Constrained 3D Gaussians for 2D Video Compression

Core Context Aware Transformers for Long Context Language Modeling

Learning Fine-Grained Representations through Textual Token Disentanglement in Composed Video Retrieval

EMMA: Empowering Multi-modal Mamba with Structural and Hierarchical Alignment

DiffPC: Diffusion-based High Perceptual Fidelity Image Compression with Semantic Refinement

Building Vision Models upon Heat Conduction

QFree-Det: Query-Free Detector with Transformer and Sequential Matching

MambaVC: Exploring Selective State Spaces for Learned Visual Compression

Core Context Aware Attention for Long Context Language Modeling

VideoUntier: Language-guided Video Feature Disentanglement