Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ICL-Router: In-Context Learned Model Representations for LLM Routing

About

Large language models (LLMs) often exhibit complementary strengths. Model routing harnesses these strengths by dynamically directing each query to the most suitable model, given a candidate model pool. However, routing performance relies on accurate model representations, and adding new models typically requires retraining, limiting scalability. To address these challenges, we propose a novel routing method using in-context vectors to represent model capabilities. The method proceeds in two stages. First, queries are embedded and projected into vectors, with a projector and LLM-based router trained to reconstruct the original queries, aligning vector representations with the router's semantic space. Second, each candidate model is profiled on a query set, and the router learns -- based on in-context vectors of query and model performance -- to predict whether each model can correctly answer new queries. Extensive experiments demonstrate that our method achieves state-of-the-art routing performance in both in-distribution and out-of-distribution tasks. Moreover, our method allows for seamless integration of new models without retraining the router. The code is available at https://github.com/lalalamdbf/ICL-Router.

Chenxu Wang, Hao Li, Yiqun Zhang, Linyao Chen, Jianhao Chen, Ping Jian, Peng Ye, Qiaosheng Zhang, Shuyue Hu• 2025

Related benchmarks

TaskDatasetResultRank
Math problem solvingMath Macro-aggregate
Pass@155.7
22
Code and Software EngineeringCode/SE Macro-aggregate
Pass@153.2
22
Knowledge retrievalKnowledge Macro-aggregate
Pass@160.8
22
Agentic Tool-useAgentic Macro-aggregate
Pass@127.3
22
Reading ComprehensionReading Macro-aggregate
Pass@147.5
22
Language Model Routing and OrchestrationBlind pool
Average Precision (p@1)49.85
16
Showing 6 of 6 rows

Other info

Follow for update