Large Language Models Can be Lazy Learners: Analyze Shortcuts in In-Context Learning
This work addresses robustness issues in in-context learning for AI researchers, though it is incremental as it builds on existing understanding of LLM behavior.
The paper investigates how large language models (LLMs) rely on shortcuts or spurious correlations in prompts for in-context learning, revealing that they are 'lazy learners' and that larger models are more likely to use these shortcuts during inference.
Large language models (LLMs) have recently shown great potential for in-context learning, where LLMs learn a new task simply by conditioning on a few input-label pairs (prompts). Despite their potential, our understanding of the factors influencing end-task performance and the robustness of in-context learning remains limited. This paper aims to bridge this knowledge gap by investigating the reliance of LLMs on shortcuts or spurious correlations within prompts. Through comprehensive experiments on classification and extraction tasks, we reveal that LLMs are "lazy learners" that tend to exploit shortcuts in prompts for downstream tasks. Additionally, we uncover a surprising finding that larger models are more likely to utilize shortcuts in prompts during inference. Our findings provide a new perspective on evaluating robustness in in-context learning and pose new challenges for detecting and mitigating the use of shortcuts in prompts.