When Many-Shot Prompting Fails: An Empirical Study of LLM Code Translation
This addresses the problem of optimizing prompting strategies for software engineers using LLMs, showing it's task-dependent and incremental in refining ICL methods.
The study tackled the assumption that many-shot prompting improves LLM performance for code translation, finding that functional correctness peaks with 5-25 examples and degrades with more, challenging the 'more is better' approach.
Large Language Models (LLMs) with vast context windows offer new avenues for in-context learning (ICL), where providing many examples ("many-shot" prompting) is often assumed to enhance performance. We investigate this assumption for the complex task of code translation. Through a large-scale empirical study of over 90,000 translations, we systematically evaluate the impact of scaling in-context examples from zero-shot to many-shot configurations of up to 625 examples, with prompts spanning from approximately 100,000 to 800,000 tokens. Our findings reveal a "many-shot paradox": while static similarity metrics may modestly improve with more examples, functional correctness consistently peaks with few-shot prompting (5-25 examples). Providing substantially more examples often degrades this crucial functional performance. This study highlights that for code translation, the quality of a few well-chosen examples outweighs sheer quantity, challenging the universal efficacy of "more is better" for ICL and underscoring the task-dependent nature of optimal prompting strategies. Our results have significant implications for effectively leveraging LLMs in software engineering.