Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204232669> ?p ?o ?g. }
- W3204232669 endingPage "11574" @default.
- W3204232669 startingPage "11566" @default.
- W3204232669 abstract "Expert-layman text style transfer technologies have the potential to improve communication between members of scientific communities and the general public. High-quality information produced by experts is often filled with difficult jargon laypeople struggle to understand. This is a particularly notable issue in the medical domain, where layman are often confused by medical text online. At present, two bottlenecks interfere with the goal of building high-quality medical expert-layman style transfer systems: a dearth of pretrained medical-domain language models spanning both expert and layman terminologies and a lack of parallel corpora for training the transfer task itself. To mitigate the first issue, we propose a novel language model (LM) pretraining task, Knowledge Base Assimilation, to synthesize pretraining data from the edges of a graph of expert- and layman-style medical terminology terms into an LM during self-supervised learning. To mitigate the second issue, we build a large-scale parallel corpus in the medical expert-layman domain using a margin-based criterion. Our experiments show that transformer-based models pretrained on knowledge base assimilation and other well-established pretraining tasks fine-tuning on our new parallel corpus leads to considerable improvement against expert-layman transfer benchmarks, gaining an average relative improvement of our human evaluation, the Overall Success Rate (OSR), by 106%." @default.
- W3204232669 created "2021-10-11" @default.
- W3204232669 creator A5029380651 @default.
- W3204232669 creator A5034254782 @default.
- W3204232669 creator A5050195037 @default.
- W3204232669 creator A5057535469 @default.
- W3204232669 date "2022-06-28" @default.
- W3204232669 modified "2023-09-26" @default.
- W3204232669 title "Self-Supervised Knowledge Assimilation for Expert-Layman Text Style Transfer" @default.
- W3204232669 cites W202246478 @default.
- W3204232669 cites W2101105183 @default.
- W3204232669 cites W2104284010 @default.
- W3204232669 cites W2134800885 @default.
- W3204232669 cites W2159583324 @default.
- W3204232669 cites W2163068962 @default.
- W3204232669 cites W2461708070 @default.
- W3204232669 cites W2467834614 @default.
- W3204232669 cites W2563351168 @default.
- W3204232669 cites W2626778328 @default.
- W3204232669 cites W2735642330 @default.
- W3204232669 cites W2739740656 @default.
- W3204232669 cites W2798389157 @default.
- W3204232669 cites W2839174743 @default.
- W3204232669 cites W2888173624 @default.
- W3204232669 cites W2905749056 @default.
- W3204232669 cites W2911489562 @default.
- W3204232669 cites W2914442349 @default.
- W3204232669 cites W2925863688 @default.
- W3204232669 cites W2937845937 @default.
- W3204232669 cites W2945440166 @default.
- W3204232669 cites W2951675429 @default.
- W3204232669 cites W2952468927 @default.
- W3204232669 cites W2953063840 @default.
- W3204232669 cites W2962739339 @default.
- W3204232669 cites W2962917899 @default.
- W3204232669 cites W2963341956 @default.
- W3204232669 cites W2963366196 @default.
- W3204232669 cites W2963602293 @default.
- W3204232669 cites W2963631950 @default.
- W3204232669 cites W2963667126 @default.
- W3204232669 cites W2964121744 @default.
- W3204232669 cites W2964321064 @default.
- W3204232669 cites W2970858854 @default.
- W3204232669 cites W2971975722 @default.
- W3204232669 cites W2973020036 @default.
- W3204232669 cites W2973154071 @default.
- W3204232669 cites W2982399380 @default.
- W3204232669 cites W2988451549 @default.
- W3204232669 cites W2996403597 @default.
- W3204232669 cites W2996888978 @default.
- W3204232669 cites W3035607998 @default.
- W3204232669 cites W3089748971 @default.
- W3204232669 cites W3099852471 @default.
- W3204232669 cites W3115753579 @default.
- W3204232669 cites W1959003973 @default.
- W3204232669 doi "https://doi.org/10.1609/aaai.v36i10.21410" @default.
- W3204232669 hasPublicationYear "2022" @default.
- W3204232669 type Work @default.
- W3204232669 sameAs 3204232669 @default.
- W3204232669 citedByCount "2" @default.
- W3204232669 countsByYear W32042326692023 @default.
- W3204232669 crossrefType "journal-article" @default.
- W3204232669 hasAuthorship W3204232669A5029380651 @default.
- W3204232669 hasAuthorship W3204232669A5034254782 @default.
- W3204232669 hasAuthorship W3204232669A5050195037 @default.
- W3204232669 hasAuthorship W3204232669A5057535469 @default.
- W3204232669 hasBestOaLocation W32042326691 @default.
- W3204232669 hasConcept C105002631 @default.
- W3204232669 hasConcept C121332964 @default.
- W3204232669 hasConcept C134306372 @default.
- W3204232669 hasConcept C138885662 @default.
- W3204232669 hasConcept C150899416 @default.
- W3204232669 hasConcept C154945302 @default.
- W3204232669 hasConcept C162324750 @default.
- W3204232669 hasConcept C165801399 @default.
- W3204232669 hasConcept C187736073 @default.
- W3204232669 hasConcept C204321447 @default.
- W3204232669 hasConcept C2780451532 @default.
- W3204232669 hasConcept C33923547 @default.
- W3204232669 hasConcept C36503486 @default.
- W3204232669 hasConcept C41008148 @default.
- W3204232669 hasConcept C41895202 @default.
- W3204232669 hasConcept C4554734 @default.
- W3204232669 hasConcept C547195049 @default.
- W3204232669 hasConcept C58328972 @default.
- W3204232669 hasConcept C62520636 @default.
- W3204232669 hasConcept C66322947 @default.
- W3204232669 hasConceptScore W3204232669C105002631 @default.
- W3204232669 hasConceptScore W3204232669C121332964 @default.
- W3204232669 hasConceptScore W3204232669C134306372 @default.
- W3204232669 hasConceptScore W3204232669C138885662 @default.
- W3204232669 hasConceptScore W3204232669C150899416 @default.
- W3204232669 hasConceptScore W3204232669C154945302 @default.
- W3204232669 hasConceptScore W3204232669C162324750 @default.
- W3204232669 hasConceptScore W3204232669C165801399 @default.
- W3204232669 hasConceptScore W3204232669C187736073 @default.
- W3204232669 hasConceptScore W3204232669C204321447 @default.
- W3204232669 hasConceptScore W3204232669C2780451532 @default.