Trade-off between Pretraining and Retrieval for AmbiEval Performance

This study presents a systematic comparison of three approaches for the analysis of mental health text using large language models (LLMs): prompt engineering, retrieval augmented generation (RAG), and fine-tuning. Using LLaMA 3, we evaluate these approaches on emotion classification and mental health condition detection tasks across two datasets. Fine-tuning achieves the highest accuracy (91\% for emotion classification, 80\% for mental health conditions) but requires substantial computational resources and large training sets, while prompt engineering and RAG offer more flexible deployment with

Research goal: What is the optimal trade-off curve between parametric knowledge from pretraining and non-parametric knowledge from retrieval for maximizing scores on the AmbiEval dataset?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.7/10.