Xianghao Kong's picture

5 2

Xianghao Kong

refkxh

·

https://refkxh.github.io/

AI & ML interests

Computer Vision

Recent Activity

authored a paper 10 days ago

Taming Flow-based I2V Models for Creative Video Editing

authored a paper 14 days ago

3D-SPS: Single-Stage 3D Visual Grounding via Referred Point Progressive Selection

authored a paper 14 days ago

Composing Concepts from Images and Videos via Concept-prompt Binding

View all activity

Organizations

upvoted a paper 15 days ago

Composing Concepts from Images and Videos via Concept-prompt Binding

Paper • 2512.09824 • Published 15 days ago • 27

upvoted a paper 25 days ago

AnyTalker: Scaling Multi-Person Talking Video Generation with Interactivity Refinement

Paper • 2511.23475 • Published 27 days ago • 41

upvoted a paper 6 months ago

Light of Normals: Unified Feature Representation for Universal Photometric Stereo

Paper • 2506.18882 • Published Jun 23 • 89

upvoted 2 papers 7 months ago

MMR-V: What's Left Unsaid? A Benchmark for Multimodal Deep Reasoning in Videos

Paper • 2506.04141 • Published Jun 4 • 29

Establishing Trustworthy LLM Evaluation via Shortcut Neuron Analysis

Paper • 2506.04142 • Published Jun 4 • 27