GFMate: Empowering Graph Foundation Models with Test-time Prompt Tuning
Not on the map. Why?
Abstract (arXiv)
Graph prompt tuning has shown great potential in graph learning by introducing trainable prompts to enhance the model performance in conventional single-domain scenarios. Recent research has extended graph prompts to improve Graph Foundation Models (GFMs) by few-shot tuning auxiliary prompts. Despite their progress, most existing methods embed source-domain information into prompts, which serve either as input to GFMs or encoded during model pre-training. Such prompt entanglement with specific source domains and GFM pre-training strategy restricts their generalisability to other domains and different GFMs. Furthermore, existing GFM prompts merely rely on few-shot tuning for adaptation, neglecting the rich information in unlabelled target domain test data. Motivated by these insights, this paper aims to empower GFMs with pre-training-agnostic test-time graph prompt tuning, named GFMate. GFMate introduces centroid and layer prompts applied after pre-training on target domains, avoiding entanglement with specific source domains and model pre-training. In addition, a test-time complementary learning objective is devised to exploit both labelled and unlabelled target domain data for effective test-time prompt tuning. Extensive experiments on 12 benchmark datasets demonstrate the superior performance and efficiency of GFMate, achieving improvements of up to 30.63%. Code is available at https://github.com/YanJiangJerry/GFMate.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 51 of its 66 references that are in the map. First posted on arXiv 14 May 2026.
Builds on
51 of 66 references.
References · 51
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
- 2017Graph Attention Networkscited by 8,923
- 2017Inductive Representation Learning on Large Graphscited by 6,320
- 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,163
- 2016Variational Graph Auto-Encoderscited by 1,382
- 2020Graph Contrastive Learning with Augmentationscited by 1,208
- 2018Deep Graph Infomaxcited by 1,106
- 2018Pitfalls of Graph Neural Network Evaluationcited by 1,007
- 2016Revisiting Semi-Supervised Learning with Graph Embeddingscited by 976
- 2020Geom-GCN: Geometric Graph Convolutional Networkscited by 826
- 2020Adaptive Universal Generalized PageRank Graph Neural Networkcited by 595
38 more
- 2019Multi-scale Attributed Node Embeddingcited by 548
- 2020Deep Graph Contrastive Representation Learningcited by 539
- 2022GraphMAE: Self-Supervised Masked Graph Autoencoderscited by 448
- 2022Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Modelscited by 382
- 2021MEMO: Test Time Robustness via Adaptation and Augmentationcited by 377
- 2020GPT-GNN: Generative Pre-Training of Graph Neural Networkscited by 303
- 2023One for All: Towards Training One Graph Model for All Classification Taskscited by 243
- 2024LLaGA: Large Language and Graph Assistantcited by 181
- 2023GraphPrompt: Unifying Pre-Training and Downstream Tasks for Graph Neural Networkscited by 167
- 2023Diverse Data Augmentation with Diffusions for Effective Test-time Prompt Tuningcited by 154
- 2023All in One: Multi-task Prompting for Graph Neural Networkscited by 148
- 2023Graph Foundation Models: Concepts, Opportunities and Challengescited by 103
- 2023Align Your Prompts: Test-Time Prompting with Distribution Alignment for Zero-Shot Generalizationcited by 101
- 2022Universal Prompt Tuning for Graph Neural Networkscited by 100
- 2021Self-Supervised Aggregation of Diverse Experts for Test-Agnostic Long-Tailed Recognitioncited by 95
- 2017Learning from Complementary Labelscited by 88
- 2024All in One and One for All: A Simple yet Effective Method towards Cross-domain Graph Pretrainingcited by 86
- 2024ZeroG: Investigating Cross-dataset Zero-shot Transferability in Graphscited by 85
- 2024OpenGraph: Towards Open Graph Foundation Modelscited by 78
- 2020Contrastive Self-supervised Learning for Graph Classificationcited by 68
- 2022Empowering Graph Representation Learning with Test-Time Graph Transformationcited by 67
- 2025SAMGPT: Text-free Graph Foundation Model for Multi-domain Pre-training and Cross-domain Adaptationcited by 62
- 2018Complementary-Label Learning for Arbitrary Losses and Modelscited by 57
- 2024GOFA: A Generative One-For-All Model for Joint Graph Language Modelingcited by 57
- 2024Text-Free Multi-domain Graph Pre-training: Toward Graph Foundation Modelscited by 42
- 2025Multi-Domain Graph Foundation Models: Robust Knowledge Transfer via Topology Alignmentcited by 42
- 2023Generalized Graph Prompt: Toward a Unification of Pre-Training and Downstream Tasks on Graphscited by 39
- 2024Fully-inductive Node Classification on Arbitrary Graphscited by 39
- 2023Test-Time Adaptation with CLIP Reward for Zero-Shot Generalization in Vision-Language Modelscited by 38
- 2025RiemannGFM: Learning a Graph Foundation Model from Riemannian Geometrycited by 35
- 2024GraphWiz: An Instruction-Following Language Model for Graph Problemscited by 32
- 2023GraphPatcher: Mitigating Degree Bias for Graph Neural Networks via Test-time Augmentationcited by 31
- 2024Non-Homophilic Graph Pre-Training and Prompt Learningcited by 31
- 2024ProG: A Graph Prompt Learning Benchmarkcited by 29
- 2022Test-Time Training for Graph Neural Networkscited by 17
- 2022GraphTTA: Test Time Adaptation on Graph Neural Networkscited by 14
- 2024Online GNN Evaluation Under Test-time Graph Distribution Shiftscited by 12
- 2025DAGPrompT: Pushing the Limits of Graph Prompting with a Distribution-aware Graph Prompt Tuning Approachcited by 11
50 most cited of 51 All 51 (JSON)
Built on by
Citing papers per year: 2026: 1.
1 paper.