Can I read A Unified Approach to Routing and Cascading for LLMs on EtoBox?

A Unified Approach to Routing and Cascading for LLMs by Dekoninck, Jasper; Baader, Maximilian; Vechev, Martin is a scholarly article available to read on EtoBox.

What is A Unified Approach to Routing and Cascading for LLMs about?

The availability of a wide range of large language models (LLMs) embedded in various agentic systems has significantly increased the potential of model selection strategies to improve the cost-performance tradeoff. Existing strategies involve either routing, where a single model is chosen per query, or cascading, which sequentially runs increasingly larger models until a satisfactory answer is found. However, current approaches face three key limitations: they (1) lack formal proofs of optimality, (2) fail to identify the conditions under which these strategies are most effective to improve the cost-performance tradeoff, and (3) are unable to combine both paradigms for further improvements. To address these issues, we first derive a novel optimal strategy for cascading and prove the optimality of an existing routing strategy. Further, we propose cascade routing, a unified framework that integrates routing and cascading into a theoretically optimal strategy. Through our analysis, we identify good quality estimators as the critical factor for the success of model selection paradigms. Finally, in our experiments, we show that cascade routing consistently outperforms the individual app

Author

Dekoninck, Jasper; Baader, Maximilian; Vechev, Martin

Published

2024

Language

EN