研究人员推出了一种新颖的图基础模型 Mochi，该模型采用元学习框架来增强任务统一性和训练效率。与依赖单独对齐步骤的先前方法不同，Mochi 在直接模拟下游评估协议的少样本（few-shot）试验上进行预训练。这种方法将训练目标与推理对齐，与现有模型相比，可带来更高的性能和显著缩短的训练时间。 AI
影响 Mochi 展示了一种更高效的训练图基础模型的方法，有望降低计算成本并加速基于图的人工智能研究。
排序理由 这是一篇描述新模型和新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2604.22031v1 Announce Type: new Abstract: 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 predic…
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 representati…