Zero-shot Entity Linking with Efficient Long Range Sequence Modeling
This addresses the problem of linking entities not seen during training for NLP researchers, though it is incremental as it builds on existing BERT-based methods.
The paper tackles zero-shot entity linking by proposing an efficient position embeddings initialization method called Embedding-repeat, which improves SOTA accuracy from 76.06% to 79.08% on a benchmark dataset and from 74.57% to 82.14% on long data.
This paper considers the problem of zero-shot entity linking, in which a link in the test time may not present in training. Following the prevailing BERT-based research efforts, we find a simple yet effective way is to expand the long-range sequence modeling. Unlike many previous methods, our method does not require expensive pre-training of BERT with long position embedding. Instead, we propose an efficient position embeddings initialization method called Embedding-repeat, which initializes larger position embeddings based on BERT-Base. On Wikia's zero-shot EL dataset, our method improves the SOTA from 76.06% to 79.08%, and for its long data, the corresponding improvement is from 74.57% to 82.14%. Our experiments suggest the effectiveness of long-range sequence modeling without retraining the BERT model.