Enhancing Function-Calling Capabilities in LLMs: Strategies for Prompt Formats, Data Integration, and Multilingual Translation

TL;DR

This paper explores strategies to improve function-calling in large language models through prompt design, data blending, a new Decision Token, reasoning techniques, and multilingual translation pipelines, enhancing accuracy and multilingual support.

Contribution

The paper introduces a Decision Token, combines instruction-following data with synthetic data, and develops a translation pipeline to advance function-calling and multilingual capabilities in LLMs.

Findings

Instruction-following data boosts accuracy and relevance detection.

Decision Token improves relevance detection with synthetic data.

Translation pipeline significantly enhances performance in Traditional Chinese.

Abstract

Large language models (LLMs) have significantly advanced autonomous agents, particularly in zero-shot tool usage, also known as function calling. This research delves into enhancing the function-calling capabilities of LLMs by exploring different approaches, including prompt formats for integrating function descriptions, blending function-calling and instruction-following data, introducing a novel Decision Token for conditional prompts, leveraging chain-of-thought reasoning, and overcoming multilingual challenges with a translation pipeline. Our key findings and contributions are as follows: (1) Instruction-following data improves both function-calling accuracy and relevance detection. (2) The use of the newly proposed Decision Token, combined with synthetic non-function-call data, enhances relevance detection. (3) A tailored translation pipeline effectively overcomes multilingual limitations, demonstrating significant improvements in Traditional Chinese. These insights highlight the potential for improved function-calling capabilities and multilingual applications in LLMs.