Liujuan Cao

FlashWorld: High-quality 3D Scene Generation within Seconds

FlexibleLLM: Making Low-Bit Quantization for Large Language Models More Flexible and Efficient

LightMotion: A Light and Tuning-free Method for Simulating Camera Motion in Video Generation

RoPK: A Head-Level Key Cache Channel Pruning Method for Efficient Long-Context LLM Inference

HGT-UCOD: A Hint-Guided Teacher Framework for Unsupervised Camouflaged Object Detection

FastVGGT: Fast Visual Geometry Transformer

Breaking the Bias: Quantifying the Attention of Industrial Anomaly Detection

Pseudo-Label Supervision in Unsupervised Industrial Anomaly Detection

FlexiReID: Adaptive Mixture of Expert for Multi-Modal Person Re-Identification

DS-VLM: Diffusion Supervision Vision Language Model

Discovering Important Experts for Mixture-of-Experts Models Pruning Through a Theoretical Perspective

Weakly-supervised 3D Referring Expression Segmentation

CTV-FAS: Compensate Texts with Visuals for Generalizable Face Anti-spoofing

Less is More: Learning Reference Knowledge Using No-Reference Image Quality Assessment