@online{Duka2609.07540,
TITLE = {{TeMo}: Temperature Modulation for Multimodal Contrastive Learning},
AUTHOR = {Duka, Dhimitrios and Schiele, Bernt and Kuehne, Hilde and Kukleva, Anna},
LANGUAGE = {eng},
URL = {https://arxiv.org/abs/2609.07540},
EPRINT = {2609.07540},
EPRINTTYPE = {arXiv},
YEAR = {2026},
ABSTRACT = {Contrastive learning approaches achieve strong performance by training models to bring similar samples closer while pushing dissimilar samples apart. A crucial component of contrastive learning is the temperature hyperparameter $$\tau$$, which controls the penalty strength applied to negative samples. However, most existing methods either fix this hyperparameter or learn a global value during training. In this paper, we introduce TeMo, Temperature Modulation framework, a similarity-based modulation approach that adaptively adjusts the temperature for each positive-negative pair according to their similarity, enabling more fine-grained multimodal contrastive learning. Our approach seamlessly integrates temperature-modulated multimodal and unimodal losses with the standard multimodal contrastive loss by gradually transitioning between them. This design allows the model to capture both coarse- and fine-grained semantics at different training stages. Extensive experiments demonstrate that each component of TeMo consistently enhances performance across diverse zero-shot retrieval and classification tasks, establishing new state-of-the-art results.},
}