The Effect of MCP Context Composition on Small Language Model Performance: A Model-Dependent Marginal Utility Analysis

Zenodo (CERN European Organization for Nuclear Research) 2026 Preprint Open access

Read full text
Authors: Jeon Hyunwoo, Kim Taesung, Kang Hyun
DOI: 10.5281/zenodo.18842478
Cited by: 2
References: 1

Abstract

A 2⁴ factorial experiment investigating how MCP (Model Context Protocol) context composition affects small language model performance. Two 7B quantized models (Qwen 2.5 7B and Llama 3.1 8B) were tested across 16 combinations of four context components (Knowledge Base, Actions, Diagnostics, Guidelines) in 10 operational scenarios (3,805 total runs). Key finding: the marginal utility of additional context components is model-dependent — Qwen benefits from full context (p

Fields of Study & Subjects

Psychometric Methodologies and Testing Topic Modeling Model-Driven Software Engineering Techniques

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
3 March 2026
Language
EN
Document type
Preprint