Exploring the Limits of Few-Shot Link Prediction in Knowledge Graphs
This work challenges existing assumptions in few-shot link prediction, providing insights for researchers developing new methods in this area by highlighting the limitations imposed by data scarcity.
This paper investigates the performance limits of few-shot link prediction in knowledge graphs, where relations have limited examples. It reveals that a simple zero-shot baseline performs surprisingly well, and that few examples fundamentally restrict models to using only coarse-grained positional information, rather than fine-grained structural details.
Real-world knowledge graphs are often characterized by low-frequency relations - a challenge that has prompted an increasing interest in few-shot link prediction methods. These methods perform link prediction for a set of new relations, unseen during training, given only a few example facts of each relation at test time. In this work, we perform a systematic study on a spectrum of models derived by generalizing the current state of the art for few-shot link prediction, with the goal of probing the limits of learning in this few-shot setting. We find that a simple zero-shot baseline - which ignores any relation-specific information - achieves surprisingly strong performance. Moreover, experiments on carefully crafted synthetic datasets show that having only a few examples of a relation fundamentally limits models from using fine-grained structural information and only allows for exploiting the coarse-grained positional information of entities. Together, our findings challenge the implicit assumptions and inductive biases of prior work and highlight new directions for research in this area.