Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385571310> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4385571310 abstract "Prompt-based methods have shown their efficacy in transferring general knowledge within pre-trained language models (PLMs) for low-resource scenarios.Typically, prompt-based methods convert downstream tasks to cloze-style problems and map all labels to verbalizers.However, when applied to zero-shot entity and relation extraction, vanilla prompt-based methods may struggle with the limited coverage of verbalizers to labels and the slow inference speed.In this work, we propose a novel Discriminate Soft Prompts (DSP) approach to take advantage of the prompt-based methods to strengthen the transmission of general knowledge.Specifically, we develop a discriminative prompt method, which reformulates zero-shot tasks into token discrimination tasks without having to construct verbalizers.Furthermore, to improve the inference speed of the prompt-based methods, we design a soft prompt co-reference strategy, which leverages soft prompts to approximately refer to the vector representation of text tokens.The experimental results show that, our model outperforms baselines on two zero-shot entity recognition datasets with higher inference speed, and obtains a 7.5% average relation F1-score improvement over previous state-of-the-art models on Wiki-ZSL and FewRel." @default.
- W4385571310 created "2023-08-05" @default.
- W4385571310 creator A5008304434 @default.
- W4385571310 creator A5018985705 @default.
- W4385571310 creator A5035959880 @default.
- W4385571310 creator A5051541195 @default.
- W4385571310 creator A5057353444 @default.
- W4385571310 creator A5060268098 @default.
- W4385571310 creator A5071891248 @default.
- W4385571310 date "2023-01-01" @default.
- W4385571310 modified "2023-09-24" @default.
- W4385571310 title "DSP: Discriminative Soft Prompts for Zero-Shot Entity and Relation Extraction" @default.
- W4385571310 doi "https://doi.org/10.18653/v1/2023.findings-acl.339" @default.
- W4385571310 hasPublicationYear "2023" @default.
- W4385571310 type Work @default.
- W4385571310 citedByCount "0" @default.
- W4385571310 crossrefType "proceedings-article" @default.
- W4385571310 hasAuthorship W4385571310A5008304434 @default.
- W4385571310 hasAuthorship W4385571310A5018985705 @default.
- W4385571310 hasAuthorship W4385571310A5035959880 @default.
- W4385571310 hasAuthorship W4385571310A5051541195 @default.
- W4385571310 hasAuthorship W4385571310A5057353444 @default.
- W4385571310 hasAuthorship W4385571310A5060268098 @default.
- W4385571310 hasAuthorship W4385571310A5071891248 @default.
- W4385571310 hasBestOaLocation W43855713101 @default.
- W4385571310 hasConcept C119857082 @default.
- W4385571310 hasConcept C124101348 @default.
- W4385571310 hasConcept C153180895 @default.
- W4385571310 hasConcept C153604712 @default.
- W4385571310 hasConcept C154945302 @default.
- W4385571310 hasConcept C17744445 @default.
- W4385571310 hasConcept C195807954 @default.
- W4385571310 hasConcept C199360897 @default.
- W4385571310 hasConcept C199539241 @default.
- W4385571310 hasConcept C204321447 @default.
- W4385571310 hasConcept C25343380 @default.
- W4385571310 hasConcept C2776214188 @default.
- W4385571310 hasConcept C2776359362 @default.
- W4385571310 hasConcept C2780801425 @default.
- W4385571310 hasConcept C28490314 @default.
- W4385571310 hasConcept C38652104 @default.
- W4385571310 hasConcept C41008148 @default.
- W4385571310 hasConcept C48145219 @default.
- W4385571310 hasConcept C94625758 @default.
- W4385571310 hasConcept C97931131 @default.
- W4385571310 hasConceptScore W4385571310C119857082 @default.
- W4385571310 hasConceptScore W4385571310C124101348 @default.
- W4385571310 hasConceptScore W4385571310C153180895 @default.
- W4385571310 hasConceptScore W4385571310C153604712 @default.
- W4385571310 hasConceptScore W4385571310C154945302 @default.
- W4385571310 hasConceptScore W4385571310C17744445 @default.
- W4385571310 hasConceptScore W4385571310C195807954 @default.
- W4385571310 hasConceptScore W4385571310C199360897 @default.
- W4385571310 hasConceptScore W4385571310C199539241 @default.
- W4385571310 hasConceptScore W4385571310C204321447 @default.
- W4385571310 hasConceptScore W4385571310C25343380 @default.
- W4385571310 hasConceptScore W4385571310C2776214188 @default.
- W4385571310 hasConceptScore W4385571310C2776359362 @default.
- W4385571310 hasConceptScore W4385571310C2780801425 @default.
- W4385571310 hasConceptScore W4385571310C28490314 @default.
- W4385571310 hasConceptScore W4385571310C38652104 @default.
- W4385571310 hasConceptScore W4385571310C41008148 @default.
- W4385571310 hasConceptScore W4385571310C48145219 @default.
- W4385571310 hasConceptScore W4385571310C94625758 @default.
- W4385571310 hasConceptScore W4385571310C97931131 @default.
- W4385571310 hasLocation W43855713101 @default.
- W4385571310 hasOpenAccess W4385571310 @default.
- W4385571310 hasPrimaryLocation W43855713101 @default.
- W4385571310 hasRelatedWork W1972656095 @default.
- W4385571310 hasRelatedWork W2024160000 @default.
- W4385571310 hasRelatedWork W2061273563 @default.
- W4385571310 hasRelatedWork W2285052147 @default.
- W4385571310 hasRelatedWork W2729514902 @default.
- W4385571310 hasRelatedWork W2743258233 @default.
- W4385571310 hasRelatedWork W2773500201 @default.
- W4385571310 hasRelatedWork W2970216048 @default.
- W4385571310 hasRelatedWork W2998168123 @default.
- W4385571310 hasRelatedWork W4287995534 @default.
- W4385571310 isParatext "false" @default.
- W4385571310 isRetracted "false" @default.
- W4385571310 workType "article" @default.