Graph Serializations for Reasoning LLMs: A Study Across Knowledge Extraction and Structural Reasoning

Key points

• The way a knowledge graph is serialized as text for inclusion into LLM context can substantially affect downstream LLM performance, with prior work reporting tens of percentage points variations across formats.1
• However, existing studies consider only a limited set of representations and predate frontier reasoning models, which now offer stronger reasoning, instruction-following, and long-context capabilities.1
• While knowledge extraction is relatively insensitive to different serialization formats, the choice of the optimal format for structural reasoning tasks significantly affects accuracy.1
• We demonstrate that the optimal representation choice depends on model, query, and task and propose to approach graph serialization as a dynamic orchestration decision.1

Sources (1)

• [1]Graph Serializations for Reasoning LLMs: A Study Across Knowledge Extraction and Structural Reasoning
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 15, 11:12 PM

  We evaluate 15 graph representation formats across nine frontier LLM configurations on ontology-based knowledge extraction and structural graph reasoning tasks.

  The way a knowledge graph is serialized as text for inclusion into LLM context can substantially affect downstream LLM performance, with prior work reporting tens of percentage points variations across formats.

Extractive summary: sentences quoted from the sources.