Meet ApplyKite Twin, the application assistant that learns with you

Meet your Twin, it learns with you

Your Twin learns with you

Launch week codes:30% off annual Pro20% off monthly Pro

Ends in00h00m00s

Xihui Liu

Assistant Professor

The University of Hong Kong

Hong Kong

Research Interests

Articles (20)

Year:

2026

Year:

2026

Year:

2026

Year:

2026

Year:

2026

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Collaborators (4)

Jianing Qiu

Assistant Professor

Mohamed Bin Zayed University of Artificial Intelligence

UNITED ARAB EMIRATES

View details

Ran Yi

Assistant Professor (tenure-track)

Shanghai Jiao Tong University

CHINA

View details

Xianhao Chen

Assistant Professor

The University of Hong Kong

HONG KONG

View details

Xuefei Ning

Research-Track Assistant Professor

Tsinghua University

CHINA

View details

ApplyKite TwinLearns as you apply

Make your next move.

Prepare your application

Review drafts before you send them.

Xihui Liu

Research Interests

Articles (20)

SJD++: Improved Speculative Jacobi Decoding for Training-Free Acceleration of Discrete Auto-Regressive Text-to-Image Generation

Optimizing Split Federated Learning With Unstable Client Participation

EgoPlan-Bench2: A Benchmark for Multimodal Large Language Model Planning in Real-World Scenarios

EgoPlan-Bench: Benchmarking Multimodal Large Language Models for Human-Level Planning

EgoSim: Egocentric World Simulator for Embodied Interaction Generation

A Survey of Reasoning with Foundation Models: Concepts, Methodologies, and Outlook

Context as Memory: Scene-Consistent Interactive Long Video Generation with Memory Retrieval

OmniPart: Part-Aware 3D Generation with Semantic Decoupling and Structural Cohesion

T2I-CompBench++: An Enhanced and Comprehensive Benchmark for Compositional Text-to-Image Generation

DreamComposer++: Empowering Diffusion Models with Multi-View Conditions for 3D Content Generation

DreamWaltz-G: Expressive 3D Gaussian Avatars from Skeleton-Guided 2D Diffusion

PredBench: Benchmarking Spatio-Temporal Prediction Across Diverse Disciplines

ScanReason: Empowering 3D Visual Grounding with Reasoning Capabilities

TC4D: Trajectory-Conditioned Text-to-4D Generation

DreamComposer: Controllable 3D Object Generation via Multi-View Conditions

EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI

Hierarchical Diffusion Autoencoders and Disentangled Image Manipulation

HumanGaussian: Text-Driven 3D Human Generation with Gaussian Splatting

Point Transformer V3: Simpler, Faster, Stronger

Shape-Guided Diffusion with Inside-Outside Attention

Collaborators (4)

Jianing Qiu

Ran Yi

Xianhao Chen

Xuefei Ning

Make your next move.