Saved in:
Bibliographic Details
Main Author: Pillai, Sneh
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.03202
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915183018901504
author Pillai, Sneh
author_facet Pillai, Sneh
contents Training vision-language models for image-text alignment typically requires large datasets to achieve robust performance. In low-data scenarios, standard contrastive learning can struggle to align modalities effectively due to overfitting and unstable training dynamics. In this paper, we propose a variance-aware loss scheduling approach that dynamically adjusts the weighting of the contrastive loss based on the statistical variability (uncertainty) in the model's alignment predictions. Using a subset of the Flickr8k image-caption dataset to simulate limited data conditions, we demonstrate that our approach improves image-text retrieval accuracy compared to a fixed-weight baseline. We also compare against other adaptive weighting strategies (using output entropy and cosine similarity spread) and find that variance-aware scheduling provides the best overall trade-off. Qualitatively, our method yields more distinct multimodal embeddings as shown by t-SNE visualizations. Moreover, in a stress test with noise-injected captions and images, the variance-guided loss proves more robust, maintaining higher recall when random perturbations are introduced. These results highlight the benefit of adaptive loss weighting for multimodal alignment in low-data regimes.
format Preprint
id arxiv_https___arxiv_org_abs_2503_03202
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Variance-Aware Loss Scheduling for Multimodal Alignment in Low-Data Settings
Pillai, Sneh
Computer Vision and Pattern Recognition
Training vision-language models for image-text alignment typically requires large datasets to achieve robust performance. In low-data scenarios, standard contrastive learning can struggle to align modalities effectively due to overfitting and unstable training dynamics. In this paper, we propose a variance-aware loss scheduling approach that dynamically adjusts the weighting of the contrastive loss based on the statistical variability (uncertainty) in the model's alignment predictions. Using a subset of the Flickr8k image-caption dataset to simulate limited data conditions, we demonstrate that our approach improves image-text retrieval accuracy compared to a fixed-weight baseline. We also compare against other adaptive weighting strategies (using output entropy and cosine similarity spread) and find that variance-aware scheduling provides the best overall trade-off. Qualitatively, our method yields more distinct multimodal embeddings as shown by t-SNE visualizations. Moreover, in a stress test with noise-injected captions and images, the variance-guided loss proves more robust, maintaining higher recall when random perturbations are introduced. These results highlight the benefit of adaptive loss weighting for multimodal alignment in low-data regimes.
title Variance-Aware Loss Scheduling for Multimodal Alignment in Low-Data Settings
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.03202