Aggregate to Adapt: Node-Centric Aggregation for Multi-Source-Free Graph Domain Adaptation

Zhen Zhang, Bingsheng He · 2025

Read Aggregate to Adapt: Node-Centric Aggregation for Multi-Source-Free Graph Domain Adaptation by Zhen Zhang, Bingsheng He (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-03033.txt.

Description

Zhen Zhang and Bingsheng He wrote this paper in two thousand twenty-five. It deals with graph domain adaptation, where a model trained on a labeled graph must work on a new graph with no labels, and the source graphs cannot be shared.

Privacy and storage rules often bar access to the source data, so only the trained source models are at hand. The authors' model, GraphATA, adapts at the level of each node rather than the whole model. It gives each node its own blend of the source models' weights, picked by that node's local setting. Tests on public data sets beat recent rivals. The design also covers both whole-model and layer-wise styles of adaptation.

Catalog record

CREATORZhen Zhang, Bingsheng He
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE63 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-03033

Browse the full library · About the Tower of Babel