Enhancing Contrastive Demonstration Selection with Semantic Diversity for Robust In-Context Machine Translation
This addresses the sensitivity of in-context learning performance to demonstration selection for machine translation, though it is incremental as it builds on existing contrastive methods.
The paper tackles the problem of demonstration selection for in-context learning in machine translation by proposing DiverseConE, which enhances contrastive selection with semantic diversity, and results show it consistently outperforms strong baselines across multiple language pairs and shot settings.
In-Context Learning (ICL) empowers large language models to perform tasks by conditioning on a few input-output examples. However, the performance of ICL is highly sensitive to the selection of these demonstrations. While existing methods focus on similarity or contrastive selection, they often overlook the importance of diversity among the chosen examples. In this paper, we propose DiverseConE (Diversity-Enhanced Contrastive Example Selection), a novel approach for demonstration selection in in-context learning for machine translation. Our method builds upon contrastive selection by incorporating a diversity enhancement step based on embedding space dissimilarity. We conduct extensive experiments on the Llama2-7b model across four language pairs (English-Chinese, Chinese-English, Russian-German, German-Russian) in 1-shot and 3-shot settings, using COMET20 and COMET22 for evaluation. Our results demonstrate that DiverseConE consistently outperforms strong baseline methods, including random selection, BM25, TopK, and a state-of-the-art contrastive selection method. Further analysis, including diversity metrics and human evaluation, validates the effectiveness of our approach and highlights the benefits of considering demonstration diversity for improved translation quality.