Mochi通过元学习对齐预训练与推理，提升图模型效率和性能。
Mochi: Aligning Pre-training and Inference for Efficient Graph Foundation Models via Meta-Learning
- 用元学习在少样本任务上预训练，使训练目标贴近实际推理场景
- 在25个真实数据集上表现优于或媲美现有模型，训练时间减少8~27倍
- 适合追求高效图模型部署的研究者和工程团队
我们提出 Mochi，一种基于元学习框架的图基础模型，以解决任务统一与训练效率问题。以往模型采用基于重构的目标（如链接预测）进行预训练，并假设可通过类原型等独立步骤将表征对齐至下游任务。我们通过合成与真实世界实验表明，该方法虽直观但存在直接影响下游性能的局限性。为克服此问题，Mochi 在模拟下游评估协议的少样本任务上进行预训练，使训练目标与推理阶段对齐，而非依赖后处理的统一步骤。结果显示，Mochi 及其更强版本 Mochi++ 在覆盖节点分类、链接预测与图分类的25个真实图数据集上达到竞争力或更优性能，且训练时间比最强基线减少8～27倍。
原文摘要 · Abstract (English)
We propose Mochi, a Graph Foundation Model that addresses task unification and training efficiency by adopting a meta-learning based training framework. Prior models pre-train with reconstruction-based objectives such as link prediction, and assume that the resulting representations can be aligned with downstream tasks through a separate unification step such as class prototypes. We demonstrate through synthetic and real-world experiments that this procedure, while simple and intuitive, has limitations that directly affect downstream task performance. To address these limitations, Mochi pre-trains on few-shot episodes that mirror the downstream evaluation protocol, aligning the training objective with inference rather than relying on a post-hoc unification step. We show that Mochi, along with its more powerful variant Mochi++, achieves competitive or superior performance compared to existing Graph Foundation Models across 25 real-world graph datasets spanning node classification, link prediction, and graph classification, while requiring 8$\sim$27 times less training time than the strongest baseline.
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。