Research map / Graph learning / Knowledge and temporal graphs
HGNet: Scalable Foundation Model for Automated Knowledge Graph Generation from Scientific Literature
Devvrat Joshi, Islem Rekik
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Multi-Scale TCQK attention (ablation) | Z-NERD · SciERC, SciER, BioRED NER | F1: 73.43 -> 78.84 (SciERC); 84.43 -> 91.05 (BioRED) | Improved |
| add Orthogonal Semantic Decomposition (ablation) | Z-NERD · SPHERE zero-shot NER | F1: 65.17 -> 73.19 (Physics ZS); 63.45 -> 72.28 (MS ZS) | Improved |
| add Orthogonal Semantic Decomposition (ablation) | Z-NERD · SciERC NER | F1: 74.39 -> 78.84 | Improved |
| add CAF loss (ablation) | HGNet · SciERC RE | Rel+ F1: 47.33 -> 53.19 | Improved |
| add CAF loss (ablation) | HGNet · SPHERE Computer Science RE | Rel+ F1: 66.50 -> 79.51 | Improved |
| add Differentiable Hierarchy Loss (ablation) | HGNet · SciERC RE | Rel+ F1: 51.68 -> 53.19 | Improved |
| add Differentiable Hierarchy Loss (ablation) | HGNet · BioRED RE | Rel+ F1: 34.31 -> 33.85 | Hurt |
| use HGNet instead of HGERE | HGNet · SciERC RE | Rel+ F1: 43.86 -> 53.19 | Improved |
| use HGNet instead of HGERE in zero-shot | HGNet · Zero-shot RE Comp. Sci. | Rel+ F1: 29.81 -> 62.60 | Improved |
| use Z-NERD instead of zero-shot LLMs | Z-NERD · SciERC NER | F1: 46.20 (llama-3.3-70b) -> 78.84 | Improved |
| use hierarchy-aware HGNet vs standard GAT on SciER peer relations | HGNet · SciER RE peer | Rel+ F1: 64.35 (HGERE) vs 67.02 (HGNet) peer; GAT 64.29 | Mixed |
Closest papers by meaning
- KGGen: Extracting Knowledge Graphs from Plain Text with Language Models (NeurIPS 2025 · similarity 0.89)
- KG-FIT: Knowledge Graph Fine-Tuning Upon Open-World Knowledge (NeurIPS 2024 · similarity 0.88)
- Select and Augment: Enhanced Dense Retrieval Knowledge Graph Augmentation (Abstract Reprint) (AAAI 2024 · similarity 0.88)
- Text2NKG: Fine-Grained N-ary Relation Extraction for N-ary relational Knowledge Graph Construction (NeurIPS 2024 · similarity 0.87)
- Learning Pattern-Based Extractors from Natural Language and Knowledge Graphs: Applying Large Language Models to Wikipedia and Linked Open Data (AAAI 2024 · similarity 0.87)
- Knowledge Graph Completion with Relation-Aware Anchor Enhancement (AAAI 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
