Privacy-preserved LLM Cascade via CoT-enhanced Policy Learning

ACL ARR 2025 February Submission432 Authors

Abstract: Large Language Models (LLMs) have demonstrated exceptional capabilities in understanding and generating natural language. However, their high deployment costs often pose a barrier to practical applications, especially. Cascading local and server models offers a promising solution to this challenge. While existing studies on LLM cascades have primarily focused on the performance-cost trade-off, real-world scenarios often involve more complex requirements. This paper introduces a novel LLM Cascade strategy with Multi-Objective Optimization, enabling LLM cascades to consider additional objectives (e.g., privacy) and better align with the specific demands of real-world applications while maintaining their original cascading abilities. Extensive experiments on three benchmarks validate the effectiveness and superiority of our approach.

Paper Type: Long

Research Area: NLP Applications

Research Area Keywords: LLM Cascade, Privacy-preserved, Policy Learning

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Publicly available software and/or pre-trained models

Languages Studied: English

Submission Number: 432