整合多个大模型优势，提升推理效果。
Harnessing Multiple Large Language Models: A Survey on LLM Ensemble
- 按推理前、中、后阶段分类多模型协同方法
- 系统梳理现有技术与评估基准
- 适合研究多模型融合的学者参考
LLM Ensemble 通过在下游推理阶段综合运用多个大语言模型（LLMs），利用各自优势来提升性能，近年来受到广泛关注。由于大模型的广泛可用性及其各具特色的强项和开箱即用特性，推动了该领域的发展。本文首次对近期 LLM Ensemble 的进展进行了系统性综述。首先提出 LLM Ensemble 的分类体系，并讨论相关研究问题；其次从‘推理前集成’、‘推理中集成’、‘推理后集成’三大类别深入划分方法并回顾各类技术；最后介绍相关基准测试与应用场景，总结现有研究并提出未来方向。论文精选列表可访问 https://github.com/junchenzhi/Awesome-LLM-Ensemble。
原文摘要 · Abstract (English)
LLM Ensemble -- which involves the comprehensive use of multiple large language models (LLMs), each aimed at handling user queries during downstream inference, to benefit from their individual strengths -- has gained substantial attention recently. The widespread availability of LLMs, coupled with their varying strengths and out-of-the-box usability, has profoundly advanced the field of LLM Ensemble. This paper presents the first systematic review of recent developments in LLM Ensemble. First, we introduce our taxonomy of LLM Ensemble and discuss several related research problems. Then, we provide a more in-depth classification of the methods under the broad categories of "ensemble-before-inference, ensemble-during-inference, ensemble-after-inference'', and review all relevant methods. Finally, we introduce related benchmarks and applications, summarize existing studies, and suggest several future research directions. A curated list of papers on LLM Ensemble is available at https://github.com/junchenzhi/Awesome-LLM-Ensemble.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。