用轻量适配器实现表格图像联合建模，节省参数且性能不降。
Parameter-Efficient Adapter Tuning for Tabular-Image Multimodal Learning
- 为表格和图像分别设计适配器，仅训练少量参数。
- 在20个数据集上性能接近全量微调，参数减少超90%。
- 适合资源有限但需多模态融合的工业落地场景。
表格-图像多模态学习旨在通过联合使用结构化表格属性与视觉数据来提升预测建模效果。尽管预训练编码器能提供强模态特异性表征，但全量微调计算成本高，而冻结编码器又限制了任务特定适应能力。我们提出表格-图像适配器（TI-Adapter），一种基于适配器的高效多模态微调框架。TI-Adapter 冻结预训练表格编码器，在提取的表格嵌入后学习一个适配器；同时对图像分支采用嵌入级与瓶颈级适配器进行优化，而非全量微调。在20个表格-图像数据集上的实验表明，TI-Adapter 在保持竞争性或更优预测性能的同时，显著减少了可训练参数量。消融研究进一步验证了适配器位置对性能与实用效率的平衡作用。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。