Establishing Performance Baselines in Fine-Tuning, Retrieval-Augmented Generation and Soft-Prompting for Non-Specialist LLM Users
This work addresses the accessibility gap for non-specialist users by establishing performance baselines with commercial tools, though it is incremental as it applies existing methods without novel algorithmic contributions.
The paper tackled the problem of making advanced LLM improvement methods accessible to non-technical users by testing fine-tuning, retrieval-augmented generation (RAG), and soft-prompting on GPT 3.5 with 100 post-September 2021 questions, finding that RAG outperformed fine-tuning and the base model, and soft prompts significantly boosted all approaches.
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.