面向表格-图像多模态学习的参数高效适配器微调
Parameter-Efficient Adapter Tuning for Tabular-Image Multimodal Learning
AI总结：
提出TI-Adapter框架，通过冻结表格编码器并添加适配器，以及图像分支的嵌入层和瓶颈层适配器，实现高效多模态微调，在20个数据集上以更少参数达到或超越全微调性能。
摘要：
表格-图像多模态学习旨在通过联合使用结构化表格属性和视觉数据来提高预测建模能力。尽管预训练编码器提供了强大的模态特定表示，但全微调可能计算成本高昂，而保持编码器冻结可能限制任务特定适应。我们提出了表格-图像适配器（TI-Adapter），一种基于模态特定适配器的微调框架，用于高效的多模态适应。TI-Adapter冻结预训练的表格编码器，并在提取的表格嵌入后学习一个适配器，同时通过嵌入级和瓶颈级适配器来适应图像分支，而不是全微调。在20个表格-图像数据集上的实验表明，TI-Adapter在使用显著更少的可训练参数的情况下，达到了与全微调相当或更好的预测性能。消融研究进一步证明了适配器放置对于平衡性能和实际效率的重要性。
英文摘要：
Tabular-image multimodal learning aims to improve predictive modeling by jointly using structured tabular attributes and visual data. Although pretrained encoders provide strong modality-specific representations, full fine-tuning can be computationally expensive, while keeping encoders frozen may limit task-specific adaptation. We propose the Tabular-Image Adapter (TI-Adapter), a modality-specific adapter-based fine-tuning framework for efficient multimodal adaptation. TI-Adapter freezes the pretrained tabular encoder and learns an adapter after the extracted tabular embedding, while adapting the image branch with embedding-level and bottleneck-level adapters instead of full fine-tuning. Experiments on 20 tabular-image datasets show that TI-Adapter achieves competitive or better predictive performance than full fine-tuning while using substantially fewer trainable parameters. Ablation studies further demonstrate the importance of adapter placement for balancing performance and practical efficiency.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
