Establishing Performance Baselines in Fine-Tuning, Retrieval-Augmented Generation and Soft-Prompting for Non-Specialist LLM Users

TL;DR

This study establishes practical performance baselines for non-expert users by comparing unmodified GPT-3.5 Turbo, a OpenAI fine-tuned variant, and a Retrieval-Augmented Generation (RAG) pipeline—each tested with and without a system prompt—on a post-2021 LayerZero dataset. Using default settings across accessible platforms, the findings show RAG outperforms fine-tuning, which in turn outperforms the unmodified model, with system prompts boosting performance across all approaches. The work emphasizes accessibility and baseline comparability, noting that RAG tools have become cost- and skill-efficient for non-experts, while highlighting the trade-offs between hallucinations and factual retrieval. Overall, the paper suggests that RAG, especially with system prompting, offers the most practical path for non-specialist users to achieve reliable Q&A from LLMs in real-world settings.

Abstract

Research into methods for improving the performance of large language models (LLMs) through fine-tuning, retrieval-augmented generation (RAG) and soft-prompting has tended to focus on the use of highly technical or high-cost techniques, making many of the newly discovered approaches comparatively inaccessible to non-technical users. In this paper we tested an unmodified version of GPT 3.5, a fine-tuned version, and the same unmodified model when given access to a vectorised RAG database, both in isolation and in combination with a basic, non-algorithmic soft prompt. In each case we tested the model's ability to answer a set of 100 questions relating primarily to events that occurred after September 2021 (the point at which GPT 3.5's training data set ends). We found that if commercial platforms are used and default settings are applied with no iteration in order to establish a baseline set of outputs, a fine-tuned model outperforms GPT 3.5 Turbo, while the RAG approach out-performed both. The application of a soft prompt significantly improved the performance of each approach.

Figures (2)

• Figure 1: Excerpt from the fine-tuning file prepared by the team.
• Figure 2: Extract from the text file used to constitute the pkl vector database used in the RAG process