paper-with-me

홈 › Papers

Task Scarcity and Label Leakage in Relational Transfer Learning

2026-03-31 · Francisco Galuppo Azevedo, Clarissa Lima Loures, Denis Oliveira Correa arxiv

Training relational foundation models requires learning representations that transfer across tasks, yet available supervision is typically limited to a small number of prediction targets per database. This task scarcity causes learned representations to encode task-specific shortcuts that degrade transfer even within the same schema, a problem we call label leakage. We study this using K-Space, a modular architecture combining frozen pretrained tabular encoders with a lightweight message-passing core. To suppress leakage, we introduce a gradient projection method that removes label-predictive directions from representation updates. On RelBench, this improves within-dataset transfer by +0.145 AUROC on average, often recovering near single-task performance. Our results suggest that limited task diversity, not just limited data, constrains relational foundation models.

📄 PDF Abstract BibTeX arXiv:2603.29914

Code (0)

등록된 구현이 없습니다.

Tasks

Transfer Learning

Similar Papers 제목 키워드 기반

OpenRFM: Dissecting Relational In-Context Learning

2026-06-03 · Zhikai Chen, Junyu Yin, Jialiang Gu, Siheng Xiong 외 arxiv

Relational Foundation Models (RFMs) promise a single pre-trained predictor that, given any relational database, returns predictions in one forward pass via relational in-context learning (ICL). Yet a substantial gap sepa…

A Comprehensive Analysis of Information Leakage in Deep Transfer Learning

2020-09-04 · Cen Chen, Bingzhe Wu, Minghui Qiu, Li Wang 외

Transfer learning is widely used for transferring knowledge from a source domain to the target domain where the labeled data is scarce. Recently, deep transfer learning has achieved remarkable progress in various applica…

Transfer Learning

Beyond Visual Evidence: Revealing and Mitigating Relational Privacy Leakage in Document MLLMs

2026-08-13 · Beining Xu, Hairui Wang, Jiaxin Wang, Changsheng Chen 외 arxiv

While the privacy risks of multimodal large language models (MLLMs) have drawn significant attention, the unique vulnerabilities of domain-specific MLLMs remain largely underexplored. Focusing on document understanding M…

Key Information Extraction

A Cross-City Federated Transfer Learning Framework: A Case Study on Urban Region Profiling

2022-05-31 · Gaode Chen, Yijun Su, Xinghua Zhang, Anmin Hu 외

Data insufficiency problems (i.e., data missing and label scarcity) caused by inadequate services and infrastructures or imbalanced development levels of cities have seriously affected the urban computing tasks in real s…

Transfer Learning

Relational Retrieval: Leveraging Known-Novel Interactions for Generalized Category Discovery

2026-05-10 · Yulin Xu, Chunqi Guo, Yuanzhen Shuai, Jianyuan Ni arxiv

In this study, we tackle Generalized Category Discovery (GCD) via a Relational Retrieval perspective, explicitly coupling labeled and unlabeled data through bidirectional knowledge transfer. While existing methods treat …