The zero-shot paradigm exploits vector-based word representations extracted
from text corpora with unsupervised methods to learn general mapping functions
from other feature spaces onto word space, where the words associated to the
nearest neighbours of the mapped vectors are used as their linguistic labels.
We show that the neighbourhoods of the mapped elements are strongly polluted by
hubs, vectors that tend to be near a high proportion of items, pushing their
correct labels down the neighbour list. After illustrating the problem
empirically, we propose a simple method to correct it by taking the proximity
distribution of potential neighbours across many mapped vectors into account.
We show that this correction leads to consistent improvements in realistic
zero-shot experiments in the cross-lingual, image labeling and image retrieval
domains.

该研究提出了一种简单的方法来解决 zero-shot 任务中邻居向量被强权向心性卡住的问题，即通过考虑多个向量中潜在邻居的相似度分布来校正错误结果。结果表明，这种校正方法在跨语言、图像标记和图像检索等领域的 zero-shot 任务中均获得了稳定的改善。