Beyond Parameter Finetuning: Test-Time Representation Refinement for Node Classification
This addresses the issue of catastrophic forgetting in test-time training for graph neural networks, offering a new paradigm with theoretical grounding and practical utility for real-world deployment.
The paper tackles the problem of performance degradation in Graph Neural Networks under out-of-distribution test scenarios by proposing TTReFT, a test-time representation fine-tuning framework that shifts adaptation from model parameters to latent representations, achieving consistent and superior performance across five benchmark datasets.
Graph Neural Networks frequently exhibit significant performance degradation in the out-of-distribution test scenario. While test-time training (TTT) offers a promising solution, existing Parameter Finetuning (PaFT) paradigm suffer from catastrophic forgetting, hindering their real-world applicability. We propose TTReFT, a novel Test-Time Representation FineTuning framework that transitions the adaptation target from model parameters to latent representations. Specifically, TTReFT achieves this through three key innovations: (1) uncertainty-guided node selection for specific interventions, (2) low-rank representation interventions that preserve pre-trained knowledge, and (3) an intervention-aware masked autoencoder that dynamically adjust masking strategy to accommodate the node selection scheme. Theoretically, we establish guarantees for TTReFT in OOD settings. Empirically, extensive experiments across five benchmark datasets demonstrate that TTReFT achieves consistent and superior performance. Our work establishes representation finetuning as a new paradigm for graph TTT, offering both theoretical grounding and immediate practical utility for real-world deployment.