This is the KG & DU Team, Official R&D entity under the 360 AI Research Institute. We’ve been focusing on "Knowledge Graph" and "Document Understanding", conducting research and developing below:

1. Knowlege Graph Construction,Represenation & NLP Application

1)🏆Competition：2022. Ranked 1st in the Competition (2022.08): Ogbl-wikikg2 leadboard

2)🏆Competition：2022. Ranked 2nd in the Competition: CCKS2022-multi-modal entity alignment leadboard

3)🔬paper: 2022. Multimodal Entity Alignment Model Based on Interactive Encoder and Knowledge Distillation,(https://sigkg.cn/ccks2022/?page_id=600)

4)🏆Competition：2022. Ranked 2nd in the Competition: CAIL2022 leadboard

5)🔬Paper: 2022. Triplere: Knowledge graph embeddings via tripled relation vectors:paper

6)🔬Paper: 2026. MMKG-RDS: Reasoning Data Synthesis via Deep Mining of Multimodal Knowledge Graphs:paper

2. Document Analysis, Layout Parser, Document Understanding & Multi-modal Large Language Models

1)🛞Opesource: 2024.Chiense oriented document layout models:360LayoutAnalysis Github, Huggingface

2)🔬Paper: 2024.FlowCE:First Multi-Dimensional Evaluation of Flowchart Comprehensionfor Multimodal Large Language Models paper, Github

3)🔬Paper: 2024.Enhancing Complex Formula Recognition with Hierarchical Detail-Focused Network paper

4)🔬Paper: 2024.First Multi-Dimensional Evaluation of Flowchart Comprehension for Multimodal Large Language Modelspaper

3. Retrieval-augmented Generation & Application

1)🔬Paper: 2024.CRUD-RAG: A Comprehensive Chinese Benchmark for Retrieval-Augmented Generation of Large Language Models: paper