超越Gemini-3-Pro:大规模LLM路由与聚合的再审视
Beyond Gemini-3-Pro: Revisiting LLM Routing and Aggregation at Scale
AI总结:
本文提出JiSi框架,通过混合路由、支持集聚合选择和自适应切换技术,实现开源LLM协作超越Gemini-3-Pro。
摘要:
大型语言模型(LLMs)已迅速发展,Gemini-3-Pro设定了新的性能里程碑。在本工作中,我们探索集体智能作为替代单一扩展的方法,并证明开源LLM的协作可以超越Gemini-3-Pro。我们首先重新审视大规模LLM的路由与聚合,并识别出三个关键瓶颈:(1)当前无训练路由器受限于基于查询的范式,仅关注文本相似性;(2)最近的聚合方法仍 largely 静态,无法为不同任务选择合适的聚合器;(3)路由与聚合的互补性仍未被充分利用。为了解决这些问题,我们引入了JiSi,一种新的框架,旨在通过三个创新释放LLM协作的全部潜力:(1)查询-响应混合路由捕捉语义信息和问题难度;(2)支持集基于的聚合器选择联合评估聚合和领域容量;(3)自适应路由-聚合切换动态利用路由和聚合的优势。在九个基准测试中的全面实验表明,JiSi通过协调十个开源LLM,仅以47%的成本超越Gemini-3-Pro,并优于主流基线。这表明集体智能代表了通往人工通用智能(AGI)的新路径。
英文摘要:
Large Language Models (LLMs) have rapidly advanced, with Gemini-3-Pro setting a new performance milestone. In this work, we explore collective intelligence as an alternative to monolithic scaling, and demonstrate that open-source LLMs' collaboration can surpass Gemini-3-Pro. We first revisit LLM routing and aggregation at scale and identify three key bottlenecks: (1) current train-free routers are limited by a query-based paradigm focusing solely on textual similarity; (2) recent aggregation methods remain largely static, failing to select appropriate aggregators for different tasks;(3) the complementarity of routing and aggregation remains underutilized. To address these problems, we introduce JiSi, a novel framework designed to release the full potential of LLMs' collaboration through three innovations: (1) Query-Response Mixed Routing capturing both semantic information and problem difficulty; (2) Support-Set-based Aggregator Selection jointly evaluating the aggregation and domain capacity of aggregators; (3) Adaptive Routing-Aggregation Switch dynamically leveraging the advantages of routing and aggregation. Comprehensive experiments on nine benchmarks demonstrate that JiSi can surpass Gemini-3-Pro with only 47% costs by orchestrating ten open-source LLMs, while outperforming mainstream baselines. It suggests that collective intelligence represents a novel path towards Artificial General Intelligence (AGI).
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。