Home › Datasets › task › Entity Disambiguation
Entity Disambiguation datasets
archive 2025-07-28
13 datasets carry the task tag "Entity Disambiguation" (the task itself: Entity Disambiguation), ordered by the archive's paper count. Page 1 of 1: 13 shown of 13. Facet routes are this site's own (the archive records the tag string, not a page).
The archive holds 12,214 dataset rows; 12,172 are listed. 6 are withheld from every listing and count here as vandalised before snapshot (6 with contact-centre spam in the title, 0 with a spam description on a row that has no homepage, no paper and no papers counted; none with more than 1 paper, 0 with a benchmark), listed in withheld.json; 1 listed row carries a vandalised description, withheld on its page. This gate never withholds a row with a homepage or a paper that resolves, and a clean description; the content rules below withhold a row whose name is spam whatever else it carries. The gate is a phrase list: these are the rows it caught, not a claim that the rest is clean. Before that gate, the site's content rules withhold 36 more rows (invite-code, gambling, travel-booking, contact-centre and similar spam in the name or on a row with nothing real behind it); they have no page and are listed in withheld.json.
Filter 51 task tags shown of 3,717, by dataset count; the full filter by modality, task and language is on /datasets
Entity Disambiguation datasets 1–13 of 13
The CoNLL dataset is a widely used resource in the field of natural language processing (NLP).
187 papers · 35 benchmarks
AIDA CoNLL-YAGO contains assignments of entities to the mentions of named entities annotated for the original CoNLL 2003 entity recognition task.
64 papers · 0 benchmarks
ACE 2004 (ACE 2004 Multilingual Training Corpus)
ACE 2004 Multilingual Training Corpus contains the complete set of English, Arabic and Chinese training data for the 2004 Automatic Content Extraction (ACE) technology evaluation.
51 papers · 6 benchmarks
A large new multilingual dataset for multilingual entity linking.
8 papers · 1 benchmark
The AQUAINT Corpus consists of newswire text data in English, drawn from three sources: the Xinhua News Service (People's Republic of China), the New York Times News Service, and the Associated Press Worldstream News Service.
6 papers · 1 benchmark
SoMeSci (Software Mentions in Scientific Articles)
Knowledge about software used in scientific investigations is important for several reasons, for instance, to enable an understanding of provenance and methods involved in data handling.
5 papers · 0 benchmarks
TAC 2010 is a dataset for summarization that consists of 44 topics, each of which is associated with a set of 10 documents.
4 papers · 1 benchmark
Hansel is a human-annotated Chinese entity linking (EL) dataset, focusing on tail entities and emerging entities: - The test set contains Few-shot (FS) and zero-shot (ZS) slices, has 10K examples and uses Wikidata as the corresponding…
2 papers · 0 benchmarks
WikiSRS is a novel dataset of similarity and relatedness judgments of paired Wikipedia entities (people, places, and organizations), as assigned by Amazon Mechanical Turk workers.
2 papers · 0 benchmarks
The Wikidata-Disamb dataset is intended to allow a clean and scalable evaluation of NED with Wikidata entries, and to be used as a reference in future research.
2 papers · 0 benchmarks
The DocRED Information Extraction (DocRED-IE) dataset extends the DocRED dataset for the Document-level Closed Information Extraction (DocIE) task.
1 paper · 6 benchmarks
ShadowLink dataset is designed to evaluate the impact of entity overshadowing on the task of entity disambiguation.
1 paper · 0 benchmarks
Dataset for our paper Disambiguation-Centric Finetuning Makes Enterprise Tool-Calling LLMs More Realistic and Less Risky which includes 5000 enterprise tools and the corresponding dialogues generated using DiaFORGE UTC data engine.
1 paper · 0 benchmarks
Paper counts and descriptions are the archive's, frozen 2025-07-28; no citation counts, no stars, no trending. Sorting by "most cited" or "newest" was a live-site feature the archive does not carry.