Order-preserving Dimension Reduction For Multimodal Semantic Embedding

Abstract

Searching for the \(k\)-nearest neighbors (KNN) in multimodal data retrieval is computationally expensive, particularly due to the inherent difficulty in comparing similarity measures across different modalities. Recent advances in multimodal machine learning address this issue by mapping data into a shared embedding space; however, the high dimensionality of these embeddings (hundreds to thousands of dimensions) presents a challenge for time-sensitive vision applications. This work proposes Order-Preserving Dimension Reduction (OPDR), aiming to reduce the dimensionality of embeddings while preserving the ranking of KNN in the lower-dimensional space. One notable component of OPDR is a new measure function to quantify KNN quality as a global metric, based on which we derive a closed-form map between target dimensionality and key contextual parameters. We have integrated OPDR with multiple state-of-the-art dimension-reduction techniques, distance functions, and embedding models; experim

Stats

• citations1
• S2 citations—
• github stars0
• HF likes0
• heat score2.26
• arxiv keygong2024order

Related papers

• TLDR: Twin Learning For Dimensionality Reduction (2021)0.00
• Dimensionality-reduction Techniques For Approximate Nearest Neighbor Search: A Survey And Evaluation (2024)0.00
• Gleanvec: Accelerating Vector Search With Minimalist Nonlinear Dimensionality Reduction (2024)0.00
• Preserving Semantic Neighborhoods For Robust Cross-modal Retrieval (2020)10.07
• Dimension-accuracy Tradeoffs In Contrastive Embeddings For Triplets, Terminals & Top-k Nearest Neighbors (2023)0.00
• Efficient Data-aware Distance Comparison Operations For High-dimensional Approximate Nearest Neighbor Search (2024)5.24
• Dimensionality Reduced Clustered Data And Order Partition And Stepwise Dimensionality Increasing Indices (2024)0.00
• Modal-aware Features For Multimodal Hashing (2019)0.00