High-dimensional dense embeddings have become central to modern Information Retrieval, but many dimensions are noisy or redundant. Recently proposed DIME (Dimension IMportance Estimation), provides query-dependent scores to identify informative components of embeddings. DIME relies on a costly grid search to select a priori a dimensionality for all the query corpus's embeddings. Our work provides a statistically grounded criterion that directly identifies the optimal set of dimensions for each query at inference time. Experiments confirm achieving parity of effectiveness and reduces embedding size by an average of $\sim50\%$ across different models and datasets at inference time.

翻译：高维稠密嵌入已成为现代信息检索的核心，但许多维度存在噪声或冗余。最近提出的DIME（维度重要性估计）通过查询依赖的评分机制来识别嵌入中的信息性成分。DIME依赖计算成本高昂的网格搜索来为所有查询语料库的嵌入预先选择维度。本研究提出一种基于统计的准则，能够在推理时直接为每个查询确定最优维度集合。实验证实，该方法在保持检索效能相当的同时，在推理阶段将不同模型和数据集的嵌入规模平均降低约50%。