根据生成置信度动态选模型，让进化型AI更省算力还准
AdaptEvolve: Improving Efficiency of Evolutionary AI Agents through Adaptive Model Selection
- 用生成置信度实时判断任务难易，动态选合适大小的LLM
- 在多个基准上平均降低37.9%推理成本，准确率保留97.5%
- 适合追求高效推理的AI系统开发者，尤其关注算力优化者
进化型智能体在推理中反复调用大语言模型（LLMs），加剧了计算效率与推理能力之间的权衡。核心问题在于：如何动态选择一个在当前生成步骤中足够强大且计算高效的模型？尽管模型级联可缓解此矛盾，但现有路由策略多依赖静态启发式或外部控制器，未显式考虑模型不确定性。本文提出AdaptEvolve：一种基于内在生成置信度估计实时可解性的多模型进化精炼框架。实验表明，基于置信度的选择在多个基准上实现更优的帕累托前沿，平均降低37.9%的总推理成本，同时保持静态大模型基线97.5%的上限准确率。代码已开源：https://github.com/raypretam/adaptive_llm_selection。
原文摘要 · Abstract (English)
Evolutionary agentic systems intensify the trade-off between computational efficiency and reasoning capability by repeatedly invoking large language models (LLMs) during inference. This setting raises a central question: how can an agent dynamically select an LLM that is sufficiently capable for the current generation step while remaining computationally efficient? While model cascades offer a practical mechanism for balancing this trade-off, existing routing strategies typically rely on static heuristics or external controllers and do not explicitly account for model uncertainty. We introduce AdaptEvolve: Adaptive LLM Selection for Multi-LLM Evolutionary Refinement within an evolutionary sequential refinement framework that leverages intrinsic generation confidence to estimate real-time solvability. Empirical results show that confidence-driven selection yields a favourable Pareto frontier, reducing total inference cost by an average of 37.9% across benchmarks while retaining 97.5% of the upper-bound accuracy of static large-model baselines. Our code is available at https://github.com/raypretam/adaptive_llm_selection.
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。