MEKER: Memory Efficient Knowledge Embedding Representation for Link Prediction and Question Answering

Chekalina, V.; Razzhigaev, Anton; Sayapin, Albert; Frolov, Evgeny; Panchenko, Alexander

doi:10.18653/v1/2022.acl-srw.27

Cited by 5 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Table 4 for SimpleQuestions-Wikidata, results for KEQA and Text2Graph are taken from MEKER [6]. They evaluate both systems on a smaller split of the SimpleQuestions-Wikidata test set.…”

Section: Discussionmentioning

confidence: 99%

“…For simple questions, KEQA [17] targets at jointly recovering the question's head entity, predicate, and tail entity representations in the KG embedding spaces and then forming a query to retrieve the answer from a KG. Text2Graph [6] uses KEQA as a base, and improves on the embedding learning model by utilising CP tensor decomposition [15]. We include both these systems in our evaluation Table 4.…”

Section: Related Workmentioning

confidence: 99%

“…For the entity labels, a BM-25-based [30] label search is performed on a database of all Wikidata entity labels, out of which top-k candidates are retrieved per entity label. For this text search we use the Elasticsearch database 6 . For our experiments, we fix k at 100.…”

Section: Entity Candidatesmentioning

confidence: 99%

See 2 more Smart Citations

GETT-QA: Graph Embedding based T2T Transformer for Knowledge Graph Question Answering

Banerjee¹,

Nair²,

Usbeck³

et al. 2023

Preprint

View full text Add to dashboard Cite

In this work, we present an end-to-end Knowledge Graph Question Answering (KGQA) system named GETT-QA. GETT-QA uses T5, a popular text-to-text pre-trained language model. The model takes a question in natural language as input and produces a simpler form of the intended SPARQL query. In the simpler form, the model does not directly produce entity and relation IDs. Instead, it produces corresponding entity and relation labels. The labels are grounded to KG entity and relation IDs in a subsequent step. To further improve the results, we instruct the model to produce a truncated version of the KG embedding for each entity. The truncated KG embedding enables a finer search for disambiguation purposes. We find that T5 is able to learn the truncated KG embeddings without any change of loss function, improving KGQA performance. As a result, we report strong results for LC-QuAD 2.0 and SimpleQuestions-Wikidata datasets on end-to-end KGQA over Wikidata.

show abstract

“…In Table 4 for SimpleQuestions-Wikidata, results for KEQA and Text2Graph are taken from MEKER [6]. They evaluate both systems on a smaller split of the SimpleQuestions-Wikidata test set.…”

Section: Discussionmentioning

confidence: 99%