Diachronic Cross-modal Embeddings
This addresses the challenge of understanding evolving multimodal information for applications like content browsing and multimodal understanding tasks, though it appears incremental as it builds on existing cross-modal embedding paradigms.
The paper tackles the problem of modeling semantic shifts in multimodal data over time by introducing a diachronic cross-modal embedding (DCM) that structures visual-textual interactions temporally, with results showing it successfully organizes instances and preserves semantic correlations at each time instant.
Understanding the semantic shifts of multimodal information is only possible with models that capture cross-modal interactions over time. Under this paradigm, a new embedding is needed that structures visual-textual interactions according to the temporal dimension, thus, preserving data's original temporal organisation. This paper introduces a novel diachronic cross-modal embedding (DCM), where cross-modal correlations are represented in embedding space, throughout the temporal dimension, preserving semantic similarity at each instant t. To achieve this, we trained a neural cross-modal architecture, under a novel ranking loss strategy, that for each multimodal instance, enforces neighbour instances' temporal alignment, through subspace structuring constraints based on a temporal alignment window. Experimental results show that our DCM embedding successfully organises instances over time. Quantitative experiments, confirm that DCM is able to preserve semantic cross-modal correlations at each instant t while also providing better alignment capabilities. Qualitative experiments unveil new ways to browse multimodal content and hint that multimodal understanding tasks can benefit from this new embedding.