Efficient Dynamic Ensembling for Multiple LLM Experts

TL;DR

This paper introduces DER, an efficient dynamic ensemble reasoning method that intelligently combines multiple LLM experts conditioned on inputs, improving performance while reducing computational costs.

Contribution

The paper proposes a novel Markov Decision Process-based framework with a reward-driven agent and a knowledge transfer prompt for effective LLM ensemble reasoning.

Findings

DER achieves better performance with fewer resources than state-of-the-art methods.

The dynamic selection of LLM experts improves task-specific accuracy.

Knowledge transfer prompts enhance the complementary use of LLMs.

Abstract

LLMs have demonstrated impressive performance across various language tasks. However, the strengths of LLMs can vary due to different architectures, model sizes, areas of training data, etc. Therefore, ensemble reasoning for the strengths of different LLM experts is critical to achieving consistent and satisfactory performance on diverse inputs across a wide range of tasks. However, existing LLM ensemble methods are either computationally intensive or incapable of leveraging complementary knowledge among LLM experts for various inputs. In this paper, we propose an efficient Dynamic Ensemble Reasoning paradigm, called DER to integrate the strengths of multiple LLM experts conditioned on dynamic inputs. Specifically, we model the LLM ensemble reasoning problem as a Markov Decision Process, wherein an agent sequentially takes inputs to request knowledge from an LLM candidate and passes the output to a subsequent LLM candidate. Moreover, we devise a reward function to train a DER-Agent to dynamically select an optimal answering route given the input questions, aiming to achieve the highest performance with as few computational resources as possible. Last, to fully transfer the expert knowledge from the prior LLMs, we develop a Knowledge Transfer Prompt that enables the subsequent LLM candidates to transfer complementary knowledge effectively. Experiments demonstrate that our method uses fewer computational resources to achieve better performance compared to state-of-the-art baselines. Code and appendix are available at https://github.com/Fhujinwu/DER