Optimizing Model Selection for Compound AI Systems
- Published
- Source
- arXiv
- Paper number
- 036
- Field
- Efficiency / Multi-model
- arXiv ID
- 2502.14815
Key points
- Compound AI systems usually use the same LLM for every module, which limits overall performance.
- There has been no systematic approach to optimizing model selection across different system components.
- We developed LLMSelector, a framework that exploits monotonic performance relationships.
- It efficiently assigns models to modules using LLM-based performance estimation.
- It implements an iterative optimization approach through sequential module nomination.
- End-to-end performance is often monotonic with respect to module-level performance.
Paper links
External research summaries. These are not HDATF publications or measured product results.