HuggingFace Daily Papers（社区热门论文）· HuggingFace Daily Papers（社区热门论文）·· 8 天前AI 评分34
SeLMRoute：面向大语言模型路由的概率语义证据框架
AI 导读
SeLMRoute 通过先提取与候选模型无关的概率语义证据、再估计候选模型表现的方式实现 LLM 路由，在 LLMRouterBench（15 个数据集、20 个候选模型、11,481 条查询）上平均准确率达 72.08%±0.45，分组五折 out-of-fold 评估为 72.64%，高于最强固定候选的 69.23%。
正文
Abstract:Large language model (LLM) routing aims to select the most suitable model for each incoming query. Most existing routers learn this decision directly from query embeddings, model representations, preference data, or clusters of similar examples. Such approaches can be effective, yet the representation used for routing rarely states what a query actually requires. We introduce SeLMRoute, a routing framework that separates the extraction of candidate-independent semantic evidence from the learning of candidate performance and the application of deployment objectives. A decision model first evaluates a set of interpretable questions about the query, such as its reasoning requirements and use of external knowledge, with each judgment retained as a probability distribution. The resulting probabilistic semantic state is used by a lightweight supervised router to estimate candidate model performance. Routing objectives are applied after performance estimation, which allows the same semantic state to support performance-oriented and cost-aware decisions. On the LLMRouterBench (15 datasets, 20 candidate models, 11,481 queries), SeLMRoute achieves an average accuracy of $72.08\% \pm 0.45$, while grouped five-fold out-of-fold evaluation reaches $72.64\%$, compared with $69.23\%$ for the strongest fixed candidate. The representation achieves the highest mean performance among the evaluated semantic, dense, lexical, and domain-level representations. In a separate 13-model performance-cost setting, SeLMRoute improves performance in all five grouped splits, with a mean PerfGain of $2.66\%$. Our code is available at this https URL.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG) |
| Cite as: | arXiv:2609.34736 [cs.AI] |
| (or arXiv:2609.34736v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.34736 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Vasileios Perifanis [view email]
[v1]
Mon, 28 Sep 2026 09:31:45 UTC (269 KB)
来源：HuggingFace Daily Papers（社区热门论文） · arxiv.org