Faster, Cheaper, Better: Multi-Objective Hyperparameter Optimization for LLM and RAG Systems
This addresses the challenge of efficiently tuning complex configurations in RAG systems for practitioners, though it is incremental as it applies existing optimization methods to a new multi-objective setting.
The paper tackled the problem of optimizing multiple objectives like cost, latency, safety, and alignment in LLM and RAG systems, finding that Bayesian optimization methods outperform baselines and achieve a superior Pareto front on two new benchmark tasks.
While Retrieval Augmented Generation (RAG) has emerged as a popular technique for improving Large Language Model (LLM) systems, it introduces a large number of choices, parameters and hyperparameters that must be made or tuned. This includes the LLM, embedding, and ranker models themselves, as well as hyperparameters governing individual RAG components. Yet, collectively optimizing the entire configuration in a RAG or LLM system remains under-explored - especially in multi-objective settings - due to intractably large solution spaces, noisy objective evaluations, and the high cost of evaluations. In this work, we introduce the first approach for multi-objective parameter optimization of cost, latency, safety and alignment over entire LLM and RAG systems. We find that Bayesian optimization methods significantly outperform baseline approaches, obtaining a superior Pareto front on two new RAG benchmark tasks. We conclude our work with important considerations for practitioners who are designing multi-objective RAG systems, highlighting nuances such as how optimal configurations may not generalize across tasks and objectives.