Guanhua Zhang, Moritz Hardt · 2026 · arXiv
DOI: 10.48550/arXiv.2610.31428 · open-access PDF
大型语言模型(LLMs)在功能和成本上有所不同,这激励了适应性地为每个查询选择模型的路由方法。这些方法依赖于性能预测:给出一个查询,估计模型是否会正确回答它。大多数路由方法训练候选模型性能的单独预测器来指导模型选择。我们的工作表明,性能预测只有在候选模型在任务上低于最佳时才超过简单的基线,或者性能预测器在候选模型上增加了显著的容量。
Machine translation (Meta M2M100); the English original follows.
Large language models (LLMs) differ in capabilities and costs, motivating routing methods that adaptively choose a model for each query. These methods rely on performance prediction: given a query, estimate whether a model will answer it correctly. Most routing methods train separate predictors of candidate models' performance to guide model selection. Our work shows that performance prediction only beats a simple baseline when the candidate model is suboptimal on the task or the performance predictor adds significant capacity relative to the candidate model. Specifically, we take a model's se
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.31428, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange