Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287888678> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4287888678 abstract "Relation Induction is a very practical task in Natural Language Processing (NLP) area. In practical application scenarios, people want to induce more entity pairs having the same relation from only a few seed entity pairs. Thus, instead of the laborious supervised setting, in this paper, we focus on the minimally-supervised setting where only a couple of seed entity pairs per relation are provided. Although the conventional relation induction methods have made some success, their performance depends heavily on the quality of word embeddings. The great success of Pre-trained Language Models, such as BERT, changes the NLP area a lot, and they are proven to be able to better capture relation knowledge. In this paper, we propose a novel method to induce relation with BERT under the minimally-supervised setting. Specifically, we firstly extract proper templates from the corpus by using the mask-prediction task in BERT to build pseudo-sentences as the context of entity pairs. Then we use BERT attention weights to better represent the pseudo-sentences. In addition, We also use the IntegratedGradient of entity pairs to iteratively select better templates further. Finally, with the high-quality pseudo-sentences, we can train a better classifier for relation induction. Experiments onGoogle Analogy Test Sets (GATS), Bigger Analogy TestSet (BATS) and DiffVec demonstrate that our proposed method achieves state-of-the-art performance." @default.
- W4287888678 created "2022-07-26" @default.
- W4287888678 creator A5004615610 @default.
- W4287888678 creator A5026310569 @default.
- W4287888678 creator A5052086909 @default.
- W4287888678 date "2022-01-01" @default.
- W4287888678 modified "2023-09-30" @default.
- W4287888678 title "Minimally-Supervised Relation Induction from Pre-trained Language Model" @default.
- W4287888678 doi "https://doi.org/10.18653/v1/2022.findings-naacl.135" @default.
- W4287888678 hasPublicationYear "2022" @default.
- W4287888678 type Work @default.
- W4287888678 citedByCount "0" @default.
- W4287888678 crossrefType "proceedings-article" @default.
- W4287888678 hasAuthorship W4287888678A5004615610 @default.
- W4287888678 hasAuthorship W4287888678A5026310569 @default.
- W4287888678 hasAuthorship W4287888678A5052086909 @default.
- W4287888678 hasBestOaLocation W42878886781 @default.
- W4287888678 hasConcept C119857082 @default.
- W4287888678 hasConcept C120665830 @default.
- W4287888678 hasConcept C121332964 @default.
- W4287888678 hasConcept C124101348 @default.
- W4287888678 hasConcept C137293760 @default.
- W4287888678 hasConcept C138885662 @default.
- W4287888678 hasConcept C151730666 @default.
- W4287888678 hasConcept C154945302 @default.
- W4287888678 hasConcept C162324750 @default.
- W4287888678 hasConcept C187736073 @default.
- W4287888678 hasConcept C192209626 @default.
- W4287888678 hasConcept C204321447 @default.
- W4287888678 hasConcept C25343380 @default.
- W4287888678 hasConcept C2779343474 @default.
- W4287888678 hasConcept C2780451532 @default.
- W4287888678 hasConcept C41008148 @default.
- W4287888678 hasConcept C41895202 @default.
- W4287888678 hasConcept C51632099 @default.
- W4287888678 hasConcept C521332185 @default.
- W4287888678 hasConcept C86803240 @default.
- W4287888678 hasConcept C95623464 @default.
- W4287888678 hasConceptScore W4287888678C119857082 @default.
- W4287888678 hasConceptScore W4287888678C120665830 @default.
- W4287888678 hasConceptScore W4287888678C121332964 @default.
- W4287888678 hasConceptScore W4287888678C124101348 @default.
- W4287888678 hasConceptScore W4287888678C137293760 @default.
- W4287888678 hasConceptScore W4287888678C138885662 @default.
- W4287888678 hasConceptScore W4287888678C151730666 @default.
- W4287888678 hasConceptScore W4287888678C154945302 @default.
- W4287888678 hasConceptScore W4287888678C162324750 @default.
- W4287888678 hasConceptScore W4287888678C187736073 @default.
- W4287888678 hasConceptScore W4287888678C192209626 @default.
- W4287888678 hasConceptScore W4287888678C204321447 @default.
- W4287888678 hasConceptScore W4287888678C25343380 @default.
- W4287888678 hasConceptScore W4287888678C2779343474 @default.
- W4287888678 hasConceptScore W4287888678C2780451532 @default.
- W4287888678 hasConceptScore W4287888678C41008148 @default.
- W4287888678 hasConceptScore W4287888678C41895202 @default.
- W4287888678 hasConceptScore W4287888678C51632099 @default.
- W4287888678 hasConceptScore W4287888678C521332185 @default.
- W4287888678 hasConceptScore W4287888678C86803240 @default.
- W4287888678 hasConceptScore W4287888678C95623464 @default.
- W4287888678 hasLocation W42878886781 @default.
- W4287888678 hasOpenAccess W4287888678 @default.
- W4287888678 hasPrimaryLocation W42878886781 @default.
- W4287888678 hasRelatedWork W1542204591 @default.
- W4287888678 hasRelatedWork W2359001871 @default.
- W4287888678 hasRelatedWork W2556319748 @default.
- W4287888678 hasRelatedWork W2900730948 @default.
- W4287888678 hasRelatedWork W2949067929 @default.
- W4287888678 hasRelatedWork W3098003361 @default.
- W4287888678 hasRelatedWork W4205288553 @default.
- W4287888678 hasRelatedWork W4205820553 @default.
- W4287888678 hasRelatedWork W4289288198 @default.
- W4287888678 hasRelatedWork W796196586 @default.
- W4287888678 isParatext "false" @default.
- W4287888678 isRetracted "false" @default.
- W4287888678 workType "article" @default.