MOLEMAN: Mention-Only Linking of Entities with a Mention Annotation Network

@inproceedings{FitzGerald2021MOLEMANML,
  title={MOLEMAN: Mention-Only Linking of Entities with a Mention Annotation Network},
  author={Nicholas FitzGerald and Jan A. Botha and Daniel Gillick and Daniel M. Bikel and Tom Kwiatkowski and Andrew McCallum},
  booktitle={ACL},
  year={2021}
}
We present an instance-based nearest neighbor approach to entity linking. In contrast to most prior entity retrieval systems which represent each entity with a single vector, we build a contextualized mention-encoder that learns to place similar mentions of the same entity closer in vector space than mentions of different entities. This approach allows all mentions of an entity to serve as “class prototypes” as inference involves retrieving from the full set of labeled entity mentions in the… 

Figures and Tables from this paper

EDIN: An End-to-end Benchmark and Pipeline for Unknown Entity Discovery and Indexing
TLDR
This work introduces the end-to-end EDIN pipeline that detects, clusters, and indexes mentions of unknown entities in context, and shows that indexing a single embedding per entity unifying the information of multiple mentions works better than indexing mentions indepen-dently.
Mention Memory: incorporating textual knowledge into Transformers through entity mention attention
TLDR
The proposed model - TOME - is a Transformer that accesses the information through internal memory layers in which each entity mention in the input passage attends to the mention memory, which enables synthesis of and reasoning over many disparate sources of information within a single Transformer model.
MENTION MEMORY : INCORPORATING TEXTUAL KNOWLEDGE INTO TRANSFORMERS THROUGH ENTITY MENTION ATTENTION
TLDR
The proposed model - TOME - is a Transformer that accesses the information through internal memory layers in which each entity mention in the input passage attends to the mention memory, which enables synthesis of and reasoning over many disparate sources of information within a single Transformer model.

References

SHOWING 1-10 OF 28 REFERENCES
Learning Dense Representations for Entity Retrieval
We show that it is feasible to perform entity linking by training a dual encoder (two-tower) model that encodes mentions and entities in the same dense vector space, where candidate entities are
Entity Linking via Dual and Cross-Attention Encoders
TLDR
The rescoring model generalizes well when trained on the larger CoNLL-2003 dataset and evaluated on TACKBP-2010, and is shown to achieve state-of-the-art results on Tack BP-2010.
Improving Entity Linking by Modeling Latent Relations between Mentions
TLDR
This work treats relations as latent variables in the neural entity-linking model so that the injected structural bias helps to explain regularities in the training data and achieves the best reported scores on the standard benchmark and substantially outperforms its relation-agnostic version.
Clustering-based Inference for Zero-Shot Biomedical Entity Linking
TLDR
This paper introduces a model in which linking decisions can be made not merely by linking to a KB entity but also by grouping multiple mentions together via clustering and jointly making linking predictions.
Entity Linking in 100 Languages
TLDR
A new formulation for multilingual entity linking, where language-specific mentions resolve to a language-agnostic Knowledge Base is proposed, where the model outperforms state-of-the-art results from a far more limited cross-lingual linking task.
Multilingual Autoregressive Entity Linking
TLDR
mGENRE is a sequence-to- sequence system for the Multilingual Entity Linking (MEL) problem—the task of resolving language-specific mentions to a multilingual Knowledge Base (KB) that treats the target language as a latent variable that is marginalized at prediction time, leading to over 50% improvements in average accuracy.
Scalable Zero-shot Entity Linking with Dense Entity Retrieval
TLDR
This paper introduces a simple and effective two-stage approach for zero-shot linking, based on fine-tuned BERT architectures, and shows that it performs well in the non-zero-shot setting, obtaining the state-of-the-art result on TACKBP-2010.
Empirical Evaluation of Pretraining Strategies for Supervised Entity Linking
In this work, we present an entity linking model which combines a Transformer architecture with large scale pretraining from Wikipedia links. Our model achieves the state-of-the-art on two commonly
Revisiting Joint Modeling of Cross-document Entity and Event Coreference Resolution
TLDR
This work jointly model entity and event coreference, and proposes a neural architecture for cross-document coreference resolution using its lexical span, surrounding context, and relation to entity (event) mentions via predicate-arguments structures.
Streaming Cross Document Entity Coreference Resolution
TLDR
This paper explores identifying coreferent entity mentions across documents in high-volume streaming text, including methods for utilizing orthographic and contextual information and shows that the approach scales to at least an order of magnitude larger data than previous reported methods.
...
...