Robust Disambiguation of Named Entities in Text

By Johannes Hoffart, Mohamed Amir Yosef, Ilaria Bordino, Hagen Fürstenau, Manfred Pinkal, M. Spaniol, Bilyana Taneva, Stefan Thater, Gerhard Weikum. 2011

1,172 citations recorded by the source corpus.

Subfields: 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 Natural Language Processing

Citations in the corpus

Few-shot Learning with Retrieval Augmented Language Models (2022, background): “We evaluate our model on the following tasks and datasets included in KILT: question answering: NaturalQuestions (Kwiatkowski et al., 2019), TriviaQA (Joshi et al., 2017) and HotpotQA (Yang et al., 2018); slot ﬁlling: Zero Shot RE (Levy et al., 2017) and T-REx (Elsahar et al., 2018); entity linking:”
Knowledge Enhanced Contextual Word Representations (2019, methodology): “The candidate selectors and priors are a combination of CrossWikis, a large, precomputed dictionary that combines statistics from Wikipedia and a web corpus (Spitkovsky and Chang, 2012), and the YAGO dictionary (Hoffart et al., 2011).”
Autoregressive Entity Retrieval (2020, methodology): “Following Kolitsas et al. (2018), we considered only mentions that have entities in the KB (i.e., Wikipedia) and we used their candidate sets with the additions of the table computed by Hoffart et al. (2011).”
CM3: A Causal Masked Multimodal Model of the Internet (2022, methodology): “For the in-domain scenario, we fine-tune CM3 on the AIDA-CoNLL dataset (Hoffart et al., 2011).”
KILT: a Benchmark for Knowledge Intensive Language Tasks (2020, methodology): “AIDA CoNLL-YAGO (Hoffart et al., 2011b) supplements the CoNLL 2003 dataset (Sang and De Meulder, 2003) with Wikipedia URL annotations for all entities using the YAGO2 system (Hoffart et al., 2011a).”
Towards Continual Knowledge Learning of Language Models (2021, background): “Configuration KILT (Petroni et al., 2021) consists of 5 different tasks and 11 datasets: OpenDomain Question Answering (Joshi et al., 2017; Kwiatkowski et al., 2019; Fan et al., 2019; Yang et al., 2018), Fact Checking (Thorne et al., 2018), Entity Linking (Hoffart et al., 2011; Guo & Barbosa, 2018),”
Multi-Task Retrieval for Knowledge-Intensive Tasks (2021, background): “Another task which stands out for being markedly different in formulation is AIDA-YAGO 2 (Hoffart et al., 2011).”
E-BERT: Efficient-Yet-Effective Entity Embeddings for BERT (2019, methodology): “We train and evaluate on AIDA, a news dataset annotated with Wikipedia URLs (Hoffart et al., 2011).”
Investigating Entity Knowledge in BERT with Simple Neural End-To-End Entity Linking (2019, methodology): “(2015) for their experiments on the CoNLL03/AIDA benchmark dataset (Hoffart et al., 2011), and due to this their experiments are comparable across studies with regards to the CG step.”
Modeling Naive Psychology of Characters in Simple Commonsense Stories (2018, background): “…of annotated resources developed on the re-lated topics of sentiment analysis (Mohammad and Turney, 2013; Deng and Wiebe, 2015), entity tracking (Hoffart et al., 2011; Weston et al., 2015), and story understanding (Goyal et al., 2010; Ouyang and McKeown, 2015; Lukin et al., 2016), our study is…”
Improving Entity Linking by Modeling Latent Relations between Mentions (2018, positive): “For in-domain scenario, we used AIDA-CoNLL dataset3 (Hoffart et al., 2011).”
DeepType: Multilingual Entity Linking by Neural Type System Evolution (2018, methodology): “[Hoffart et al. 2011] Hoffart, J.; Yosef, M. A.; Bordino, I.; Fürstenau, H.; Pinkal, M.; Spaniol, M.; Taneva, B.; Thater, S.; and Weikum, G. 2011.”
Pre-training Text-to-Text Transformers for Concept-centric Common Sense (2020, methodology): “We use AIDA CoNLL-YAGO (AY2) (Hoffart et al., 2011) which supplements the CoNLL 2003 (Tjong Kim Sang & De Meulder, 2003) with Wikipedia URL annotations for all entities.”
Knowledge graph refinement: A survey of approaches and evaluation methods (2016, methodology): “Those are most often used together with DBpedia, so that there are already links between the entities and the corpus of background knowledge, i.e., no Named Entity Recognition has to be performed, in contrast to the distant supervision approaches discussed above.”
Entity Linking meets Word Sense Disambiguation: a Unified Approach (2014, methodology): “• AIDA-CoNLL 6 (Hoffart et al., 2011), which consists of 1392 English articles, for a total of roughly 35 K named entity mentions annotated with YAGO concepts separated in development, training and test sets.”