The past decade has witnessed rapid advancements in cross-modal retrieval, with significant progress made in accurately measuring the similarity between cross-modal pairs. However, the persistent hubness problem, a phenomenon where a small number of targets frequently appear as nearest neighbors to numerous queries, continues to hinder the precision of similarity measurements. Despite several proposed methods to reduce hubness, their underlying mechanisms remain poorly understood. To bridge this gap, we analyze the widely-adopted Inverted Softmax approach and demonstrate its effectiveness in balancing target probabilities during retrieval. Building on these insights, we propose a probability-balancing framework for more effective hubness reduction. We contend that balancing target probabilities alone is inadequate and, therefore, extend the framework to balance both query and target probabilities by introducing Sinkhorn Normalization (SN). Notably, we extend SN to scenarios where the true query distribution is unknown, showing that current methods, which rely solely on a query bank to estimate target hubness, produce suboptimal results due to a significant distributional gap between the query bank and targets. To mitigate this issue, we introduce Dual Bank Sinkhorn Normalization (DBSN), incorporating a corresponding target bank alongside the query bank to narrow this distributional gap. Our comprehensive evaluation across various cross-modal retrieval tasks, including image-text retrieval, video-text retrieval, and audio-text retrieval, demonstrates consistent performance improvements, validating the effectiveness of both SN and DBSN. All codes are publicly available at https://github.com/ppanzx/DBSN.

Current state-of-the-art image-text matching methods implicitly align the visual-semantic fragments, like regions in images and words in sentences, and adopt cross-attention mechanism to discover fine-grained cross-modal semantic correspondence. However, the cross-attention mechanism may bring redundant or irrelevant region-word alignments, degenerating retrieval accuracy and limiting efficiency. Although many researchers have made progress in mining meaningful alignments and thus improving accuracy, the problem of poor efficiency remains unresolved. In this work, we propose to learn fine-grained image-text matching from the perspective of information coding. Specifically, we suggest a coding framework to explain the fragments aligning process, which provides a novel view to reexamine the cross-attention mechanism and analyze the problem of redundant alignments. Based on this framework, a Cross-modal Hard Aligning Network (CHAN) is designed, which comprehensively exploits the most relevant region-word pairs and eliminates all other alignments. Extensive experiments conducted on two public datasets, MS-COCO and Flickr30K, verify that the relevance of the most associated word-region pairs is discriminative enough as an indicator of the image-text similarity, with superior accuracy and efficiency over the state-of-the-art approaches on the bidirectional image and text retrieval tasks. Our code will be available at https://github.com/ppanzx/CHAN.

Abstract Triplet loss is widely used as the objective function in image‐text retrieval tasks. However, as all the triplets are treated equally, triplet loss has a bottleneck problem of slow convergence and other unsatisfactory performances. In this article, we propose solutions by appropriately weighting triplets according to the relative similarities among the training samples. Specifically, we present three weighting functions to assign an appropriate weight for the selected informative triplets to accelerate the convergence. We evaluate our approach on two widely used benchmark datasets: Flickr30k and MSCOCO, with results outperforming the previous methods, which demonstrates its superiority.