G2CP: A Graph-Grounded Communication Protocol for Verifiable and Efficient Multi-Agent Reasoning

TL;DR

G2CP introduces a structured graph-based communication protocol for multi-agent systems, significantly reducing token usage, enhancing accuracy, and providing verifiable reasoning, thereby overcoming issues of semantic drift and hallucinations in natural language exchanges.

Contribution

This paper presents G2CP, a novel graph-grounded communication protocol that replaces free-text messages with explicit graph operations, enabling verifiable and efficient multi-agent reasoning.

Findings

Reduces inter-agent communication tokens by 73%.

Improves task accuracy by 34% over free-text methods.

Eliminates cascading hallucinations and provides auditable reasoning chains.

Abstract

Multi-agent systems powered by Large Language Models face a critical challenge: agents communicate through natural language, leading to semantic drift, hallucination propagation, and inefficient token consumption. We propose G2CP (Graph-Grounded Communication Protocol), a structured agent communication language where messages are graph operations rather than free text. Agents exchange explicit traversal commands, subgraph fragments, and update operations over a shared knowledge graph, enabling verifiable reasoning traces and eliminating ambiguity. We validate G2CP within an industrial knowledge management system where specialized agents (Diagnostic, Procedural, Synthesis, and Ingestion) coordinate to answer complex queries. Experimental results on 500 industrial scenarios and 21 real-world maintenance cases show that G2CP reduces inter-agent communication tokens by 73%, improves task completion accuracy by 34% over free-text baselines, eliminates cascading hallucinations, and produces fully auditable reasoning chains. G2CP represents a fundamental shift from linguistic to structural communication in multi-agent systems, with implications for any domain requiring precise agent coordination. Code, data, and evaluation scripts are publicly available.