Improving Entity Linking by Modeling Latent Relations between Mentions

By Phong Le, Ivan Titov. 2018

227 citations recorded by the source corpus.

Entity linking involves aligning textual mentions of named entities to their corresponding entries in a knowledge base. Entity linking systems often exploit relations between textual mentions in a document (e.g., coreference) to decide if the linking decisions are compatible. Unlike previous approaches, which relied on supervised systems or heuristics to predict these relations, we treat relations as latent variables in our neural entity-linking model. We induce the relations without any supervision while optimizing the entity-linking system in an end-to-end fashion. Our multi-relational model achieves the best reported scores on the standard benchmark (AIDA-CoNLL) and substantially outperforms its relation-agnostic version. Its training also converges much faster, suggesting that the injected structural bias helps to explain regularities in the training data.

Subfields: Artificial Intelligence, Management Science and Operations Research, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

In Natural Language Processing

Citations in the corpus

Autoregressive Entity Retrieval (2020, methodology): “Entity Disambiguation (ED) We reproduce the setting of Le & Titov (2018) using the same candidate sets, in-domain and out-of-domain datasets, and evaluating using the InKB micro-F 1 .”
CM3: A Causal Masked Multimodal Model of the Internet (2022, methodology): “We reproduce the evaluation setting described by De Cao et al. (2020) and Le & Titov (2018) using the same candidate sets, datasets and evaluating using the InKB micro-F1 metric.”
Entity-aware ELMo: Learning Contextual Entity Representation for Entity Disambiguation (2019, methodology): “We follow the same setup as our base-lines (Ganea and Hofmann, 2017; Le and Titov, 2018) and use the same train, test, validation splits, the same candidate sets and the prior feature values p ( e | m ) .”
Joint Entity Linking with Deep Reinforcement Learning (2019, methodology): “ learning are presented in Algorithm 1. 3 EXPERIMENT In order to evaluate the e‡ectiveness of our method, we train the RLEL model and validate it on a series of popular datasets that are also used by [14, 37]. To avoid over•−ing with one dataset, we use both AIDA-Train and Wikipedia data in the trai”
Learning Dynamic Context Augmentation for Global Entity Linking (2019, methodology): “Another common practice is to use Loopy Belief Propagation for inference (Ganea and Hofmann, 2017; Le and Titov, 2018).”
Unified Named Entity Recognition as Word-Word Relation Classification (2021, background): “…community, due to its wide variety of knowledge-based applications, e.g., relation extraction (Wei et al. 2020; Li et al. 2021b), entity linking (Le and Titov 2018; Hou et al. 2020), etc. Studies of NER have gradually evolved initially from the ﬂat NER (Lample et al. 2016; Strubell et al.…”
Improving Entity Linking by Modeling Latent Entity Type Information (2020, methodology): “The new local context score function considering explicit type information is deﬁned as: 1 and Table 2, in the oracle setting, the best model outperforms all the state-of-the-art entity linking models by a large margin, even surpass Le and Titov (2018) by 3.28 F1 points on AIDA-CoNLL test set.”
A Comprehensive Survey on Automatic Knowledge Graph Construction (2023, background): “Le and Titov [14] model the latent relations between mentions in the context for embedding, utilizing mention-wise and relation-wise normalization to score pair-wise coherence score function.”
Joint Learning of Named Entity Recognition and Entity Linking (2019, background): “, 2018) or by allying local and global features, using information about the neighbouring mentions and respective entities (Le and Titov, 2018; Cao et al., 2018; Yang et al., 2018).”
REL: An Entity Linker Standing on the Shoulders of Giants (2020, methodology): “The first one, denoted as REL (2014), is based on the original implementation of [15] for ED.”
Zero-Shot Entity Linking by Reading Entity Descriptions (2019, background): “…EL Recent work has also generalized to a cross-domain setting, linking entity mentions in different types of text, such as blogposts and news articles to the Wikipedia KB, while only using labeled mentions in Wikipedia for training (e.g., Gupta et al. (2017); Le and Titov (2018), inter alia ).”
End-to-End Neural Entity Linking (2018, background): “(2016); Ganea and Hofmann (2017); Le and Titov (2018); Yang et al. (2018); Radhakrishnan et al. (2018) have established state-of-the-art results, outperforming engineered features based models.”
Boosting Entity Linking Performance by Leveraging Unlabeled Documents (2019, methodology): “1% F1 (Le and Titov, 2018), whereas the best previous out-of-domain score is only 85.”