# Learn to Not Link: Exploring NIL Prediction in Entity Linking

Source: [Learn to Not Link: Exploring NIL Prediction in Entity Linking](<https://www.sota2.com/research/paper/68fa6f7c1e209131cea45e9f>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T04:30:46.200Z

Authors: Fangwei Zhu, Jifan Yu, Hailong Jin, Juanzi Li, Lei Hou, Zhifang Sui

Published: 2023-05-25T05:12:33.000Z

## Abstract

Entity linking models have achieved significant success via utilizing pretrained language models to capture semantic features. However, the NIL prediction problem, which aims to identify mentions without a corresponding entity in the knowledge base, has received insufficient attention. We categorize mentions linking to NIL into Missing Entity and Non-Entity Phrase, and propose an entity linking dataset NEL that focuses on the NIL prediction problem. NEL takes ambiguous entities as seeds, collects relevant mention context in the Wikipedia corpus, and ensures the presence of mentions linking to NIL by human annotation and entity masking. We conduct a series of experiments with the widely used bi-encoder and cross-encoder entity linking models, results show that both types of NIL mentions in training data have a significant influence on the accuracy of NIL prediction. Our code and dataset can be accessed at https://github.com/solitaryzero/NIL\_EL

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Entity Linking on NEL](<https://www.sota2.com/research/sota/entity-linking-on-nel.md>): Entity Linking; NEL; Non-NAC: 86.97.

- [Entity Linking on MSNBC (OAC)](<https://www.sota2.com/research/sota/entity-linking-on-msnbc-oac.md>): Entity Linking; MSNBC; OAC: 89.7.

- [Entity Linking on WNED-WIKI](<https://www.sota2.com/research/sota/entity-linking-on-wned-wiki.md>): Entity Linking; WNED-WIKI; OAC: 72.43.

- [Entity Linking on AIDA w/ NIL (testb)](<https://www.sota2.com/research/sota/entity-linking-on-aida-w-nil-testb.md>): Entity Linking; AIDA w/ NIL (testb); Non-NAC Accuracy: 84.42.

- [Entity Linking on AIDA w/o NIL (testb)](<https://www.sota2.com/research/sota/entity-linking-on-aida-w-o-nil-testb.md>): Entity Linking; AIDA w/o NIL (testb); OAC: 88.16.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2305.15725>)

[PDF](<https://arxiv.org/pdf/2305.15725.pdf>)

[Implementation](<https://github.com/solitaryzero/NIL_EL>)