Graph Neural Prompting (GNP)
First page
Paper summary
A plug-and-play method that injects knowledge-graph information into frozen pretrained LLMs.
Ask this paper
01
KG-to-embedding bridge: Uses a graph neural network to encode relevant knowledge-graph subgraphs into a soft prompt embedding that conditions the LLM.
02
Frozen-LLM compatible: Works with frozen pretrained LLMs without requiring any fine-tuning, making it cheap to adopt.
03
Commonsense gains: Improves performance on commonsense QA benchmarks where structured knowledge-graph information is known to help.
04
Modular extensibility: The GNN-encoded soft-prompt pattern generalizes beyond KGs to any structured input that can be encoded into embeddings.