Liang Zhao

16

Papers

301

Total Citations

1

Affiliations

Affiliations

Megvii

Papers (16)

DreamLLM: Synergistic Multimodal Comprehension and Creation

Open Vision Reasoner: Transferring Linguistic Cognitive Behavior for Visual Reasoning

Perception in Reflection

Incomplete and Unpaired Multi-View Graph Clustering with Cross-View Feature Fusion

MAGI: Multi-Annotated Explanation-Guided Learning

Background-Insensitive Scene Text Recognition with Text Semantic Segmentation

JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation

Continuous Domain Generalization

Occlusion Aware Unsupervised Learning of Optical Flow

Task-Specific Inconsistency Alignment for Domain Adaptive Object Detection

Representation Learning on Spatial Networks

Multi-objective Deep Data Generation with Correlated Property Control

Deep Generative Model for Periodic Graphs

Curriculum Learning for Graph Neural Networks: Which Edges Should We Learn First

Theoretical Properties for Neural Networks with Weight Matrices of Low Displacement Rank

Learning to Explore via Meta-Policy Gradient