AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Commonsense Knowledge Base Completion with Structural and Semantic Context

Chaitanya Malaviya, Chandra Bhagavatula, Antoine Bosselut, Yejin Choi

AAAI 2020 · Graph learning › Knowledge and temporal graphs · approach: Inductive knowledge graph completion · 140 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use BERT node representations instead of ConvTransE's learned node embeddingsConvTransE · CN-100K, subgraph samplingMRR: 18.68 -> 49.56Improved
Use BERT node representations instead of ConvTransE's learned node embeddingsConvTransE · ATOMIC, subgraph samplingMRR: 12.94 -> 12.33Hurt
Add graph convolutional node representationsConvTransE · CN-100K, subgraph samplingMRR: 18.68 -> 29.80Improved
Add graph convolutional node representationsConvTransE · ATOMIC, subgraph samplingMRR: 12.94 -> 13.12Improved
Add semantic-similarity links (ablation)GCN + ConvTransE · CN-100K, subgraph samplingMRR: 29.80 -> 30.03Improved
Add semantic-similarity links (ablation)GCN + ConvTransE · ATOMIC, subgraph samplingMRR: 13.12 -> 13.88Improved
Fuse GCN representations with BERT representations (ablation)ConvTransE · CN-100K, subgraph samplingMRR: 49.56 -> 50.38Improved
Fuse GCN representations with BERT representations (ablation)ConvTransE · ATOMIC, subgraph samplingMRR: 12.33 -> 10.8Hurt
Add semantic-similarity links to the fused GCN+BERT model (ablation)GCN + BERT + ConvTransE · CN-100K, subgraph samplingMRR: 50.38 -> 51.11Improved
Add semantic-similarity links to the fused GCN+BERT model (ablation)GCN + BERT + ConvTransE · ATOMIC, subgraph samplingMRR: 10.8 -> 10.33Hurt
Use graph embeddings alongside BERT representations (ablation)ConvTransE · CN-100K, full graph trainingMRR: 52.25 -> 50.80Hurt
Add semantic-similarity links to GCN+BERT (ablation)ConvTransE · CN-100K, full graph trainingMRR: 50.80 -> 50.17Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy