[ICLR 2026] The role of diversity in in-context learning for large language models

Wenyang Xiao, Haoyu Zhao, Lingxiao Huang

Keywords: Data selection, Diversity, In-context Learning

TL;DR: We systematically investigate the role of diversity in in-context example selection for large language models, and show that diversity-aware selection methods improve performance on complex tasks and enhance robustness to out-of-distribution queries.

Abstract: In-context learning (ICL) is a crucial capability of current large language models (LLMs), where the selection of examples plays a key role in performance. While most existing approaches focus on selecting the most similar examples to the query, the impact of diversity in example selection remains underexplored. We systematically investigate the role of diversity in in-context example selection through experiments across a range of tasks, from sentiment classification to more challenging math and code problems. Experiments on Llama-3.1, Gemma-2, and Mistral-v0.3 families of models show that diversity-aware selection methods im9 prove performance, particularly on complex tasks like math and code, and enhance robustness to out-of-distribution queries. To support these findings, we introduce a theoretical framework that explains the benefits of incorporating diversity in in-context example selection.

Primary Area: foundation or frontier models, including LLMs

Submission Number: 12909