影响力指数

45.74/100

前 10.5%

全站排名 #6,775

发表论文11 篇

平均评分4.3

年均产出11.0 篇/年

Xiaojuan Qi

Assistant Professor@University of Hong Kong·中国香港·OpenReview

研究方向

Open-world Learning · 3D Reconstruction and Generation · Segmentation · 2D and 3D Scene Understanding

6.0

Easier Painting Than Thinking: Can Text-to-Image Models Set the Stage, but Not Direct the Play?

ICLR 2026Poster

6.0

Efficient Prediction of Large Protein Complexes via Subunit-Guided Hierarchical Refinement

ICLR 2026Poster

4.0

DBellQuant: Breaking the Bell with Double-Bell Transformation for LLM Post Training Binarization

ICLR 2026Rejected

通讯

4.0

Can 3D Vision-Language Models Truly Understand Natural Language?

ICLR 2026Withdrawn

3.5

Absorbing Quantization Error by Deformable Noise Scheduler for Diffusion Models

ICLR 2026Rejected

通讯

3.3

Generative Distribution Distillation

ICLR 2026Withdrawn

2.5

Fast to Train, Fast to Sample: Stable Velocity for Flow Matching

ICLR 2026Rejected

2.5

Multi-Scale Group Relative Policy Optimization for Large Language Models

合作者 (20)

Xiaojuan Qi

Easier Painting Than Thinking: Can Text-to-Image Models Set the Stage, but Not Direct the Play?

QeRL: Beyond Efficiency - Quantization-enhanced Reinforcement Learning for LLMs

LiFR-Seg: Anytime High-Frame-Rate Segmentation via Event-Guided Propagation

Fast Data Mixture Optimization via Gradient Descent

Efficient Prediction of Large Protein Complexes via Subunit-Guided Hierarchical Refinement

DBellQuant: Breaking the Bell with Double-Bell Transformation for LLM Post Training Binarization

Can 3D Vision-Language Models Truly Understand Natural Language?

Absorbing Quantization Error by Deformable Noise Scheduler for Diffusion Models

Generative Distribution Distillation

Fast to Train, Fast to Sample: Stable Velocity for Flow Matching

Multi-Scale Group Relative Policy Optimization for Large Language Models