# Improving Entity Disambiguation by Reasoning over a Knowledge Base

arXiv: https://arxiv.org/abs/2207.04106 · HTML: https://mlsanity.com/paper/2207.04106/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 45 references that are in the map.
- First posted on arXiv 8 Jul 2022.

## Builds on (most cited first, 22 of 22)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Autoregressive Entity Retrieval](https://mlsanity.com/paper/2010.00904/) (2020)
- [DocRED: A Large-Scale Document-Level Relation Extraction Dataset](https://mlsanity.com/paper/1906.06127/) (2019)
- [Higher-order Coreference Resolution with Coarse-to-fine Inference](https://mlsanity.com/paper/1804.05392/) (2018)
- [Zero-Shot Entity Linking by Reading Entity Descriptions](https://mlsanity.com/paper/1906.07348/) (2019)
- [Scalable Zero-shot Entity Linking with Dense Entity Retrieval](https://mlsanity.com/paper/1911.03814/) (2019)
- [Deep Joint Entity Disambiguation with Local Neural Attention](https://mlsanity.com/paper/1704.04920/) (2017)
- [Entity Structure Within and Throughout: Modeling Mention Dependencies for Document-Level Relation Extraction](https://mlsanity.com/paper/2102.10249/) (2021)
- [REL: An Entity Linker Standing on the Shoulders of Giants](https://mlsanity.com/paper/2006.01969/) (2020)
- [Improving Entity Linking by Modeling Latent Relations between Mentions](https://mlsanity.com/paper/1804.10637/) (2018)
- [DeepType: Multilingual Entity Linking by Neural Type System Evolution](https://mlsanity.com/paper/1802.01021/) (2018)
- [Fine-Grained Entity Typing for Domain Independent Entity Linking](https://mlsanity.com/paper/1909.05780/) (2019)
- [Scalable Neural Methods for Reasoning With a Symbolic Knowledge Base](https://mlsanity.com/paper/2002.06115/) (2020)
- [Joint Entity Linking with Deep Reinforcement Learning](https://mlsanity.com/paper/1902.00330/) (2019)
- [Bootleg: Chasing the Tail with Self-Supervised Named Entity Disambiguation](https://mlsanity.com/paper/2010.10363/) (2020)
- [Evaluating the Impact of Knowledge Graph Context on Entity Disambiguation Models](https://mlsanity.com/paper/2008.05190/) (2020)
- [Learning Dynamic Context Augmentation for Global Entity Linking](https://mlsanity.com/paper/1909.02117/) (2019)
- [Collective Entity Disambiguation with Structured Gradient Tree Boosting](https://mlsanity.com/paper/1802.10229/) (2018)
- [Boosting Entity Linking Performance by Leveraging Unlabeled Documents](https://mlsanity.com/paper/1906.01250/) (2019)
- [Robustness Evaluation of Entity Disambiguation Using Prior Probes:the Case of Entity Overshadowing](https://mlsanity.com/paper/2108.10949/) (2021)

## Built on by (most cited first, 7 of 7)

- [AMELI: Enhancing Multimodal Entity Linking with Fine-Grained Attributes](https://mlsanity.com/paper/2305.14725/) (2023)
- [Improving Toponym Resolution with Better Candidate Generation, Transformer-based Reranking, and Two-Stage Resolution](https://mlsanity.com/paper/2305.11315/) (2023)
- [Knowledge Graphs for Enhancing Large Language Models in Entity Disambiguation](https://mlsanity.com/paper/2505.02737/) (2025)
- [Span-Oriented Information Extraction -- A Unifying Perspective on Information Extraction](https://mlsanity.com/paper/2403.15453/) (2024)
- [REXEL: An End-to-end Model for Document-Level Relation Extraction and Entity Linking](https://mlsanity.com/paper/2404.12788/) (2024)
- [ATLAS: Article Tracking, Linking, and Analysis of Swedish Encyclopedias](https://mlsanity.com/paper/2605.02466/) (2026)
- [KARAMEL: Knowledge-Aware Ranking for Multimodal Entity Linking in the Arts via Heterogeneous Graphs](https://mlsanity.com/paper/2610.16502/) (2026)

## Abstract (arXiv)

Recent work in entity disambiguation (ED) has typically neglected structured knowledge base (KB) facts, and instead relied on a limited subset of KB information, such as entity descriptions or types. This limits the range of contexts in which entities can be disambiguated. To allow the use of all KB facts, as well as descriptions and types, we introduce an ED model which links entities by reasoning over a symbolic knowledge base in a fully differentiable fashion. Our model surpasses state-of-the-art baselines on six well-established ED datasets by 1.3 F1 on average. By allowing access to all KB information, our model is less reliant on popularity-based entity priors, and improves performance on the challenging ShadowLink dataset (which emphasises infrequent and ambiguous entities) by 12.7 F1.