Knowledge Graph Prompting Sparks Graph of Thoughts in LLMs

Created in August 29, 2023

2023   ·   llm   reasoning   ·   paper-review

This paper connects knowledge graph (KG) prompting with graph-of-thought (GoT) reasoning in LLMs. Rather than prompting models with flat text, the approach encodes relational structure from knowledge graphs as explicit reasoning scaffolds, allowing LLMs to follow graph-structured inference chains rather than linear chains of thought.

The work bridges structured knowledge representation and the emerging line of graph-structured prompting methods (extending beyond CoT and ToT).



Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • The Expressive Power of Transformers with Chain of Thought
  • Landscape of Thoughts — Visualizing Where LLM Reasoning Actually Goes
  • Magellan — Guided MCTS for Escaping the Gravity Wells of LLM Creativity
  • PriorZero — Injecting LLM Priors into MuZero-Style World Models at the MCTS Root
  • SuperThoughts — Reasoning Tokens in Superposition