GreaseLM
A classic knowledge-graph-augmented language model that integrates graph neural networks with pre-trained language models for question-answering tasks. It serves as an important baseline model in the knowledge-graph-enhanced question-answering research area.
Timeline (last 90 days)
VisKG-LM achieves improvements of 1.2, 0.8, and 4.3 points over the GreaseLM baseline on the three benchmarks CommonsenseQA, OpenBookQA, and MedQA-USMLE, respectively.
GreaseLM is a classic baseline method in the knowledge graph QA field, whose core idea is to perform bidirectional fusion with a GNN at every layer of the language model.
All Facts (2)
VisKG-LM achieves improvements of 1.2, 0.8, and 4.3 points over the GreaseLM baseline on the three benchmarks CommonsenseQA, OpenBookQA, and MedQA-USMLE, respectively.
50%UnverifiedGreaseLM is a classic baseline method in the knowledge graph QA field, whose core idea is to perform bidirectional fusion with a GNN at every layer of the language model.
50%