Matches in SemOpenAlex for { <https://semopenalex.org/work/W3035331128> ?p ?o ?g. }
- W3035331128 abstract "We propose a simple data augmentation protocol aimed at providing a compositional inductive bias in conditional and unconditional sequence models. Under this protocol, synthetic training examples are constructed by taking real training examples and replacing (possibly discontinuous) fragments with other fragments that appear in at least one similar environment. The protocol is model-agnostic and useful for a variety of tasks. Applied to neural sequence-to-sequence models, it reduces error rate by as much as 87% on diagnostic tasks from the SCAN dataset and 16% on a semantic parsing task. Applied to n-gram language models, it reduces perplexity by roughly 1% on small corpora in several languages." @default.
- W3035331128 created "2020-06-19" @default.
- W3035331128 creator A5070829652 @default.
- W3035331128 date "2020-01-01" @default.
- W3035331128 modified "2023-10-17" @default.
- W3035331128 title "Good-Enough Compositional Data Augmentation" @default.
- W3035331128 cites W1493576607 @default.
- W3035331128 cites W1496189301 @default.
- W3035331128 cites W1499506821 @default.
- W3035331128 cites W1538261701 @default.
- W3035331128 cites W1602017060 @default.
- W3035331128 cites W2049221532 @default.
- W3035331128 cites W2064675550 @default.
- W3035331128 cites W2075201173 @default.
- W3035331128 cites W2075322664 @default.
- W3035331128 cites W2129882630 @default.
- W3035331128 cites W2130942839 @default.
- W3035331128 cites W2134800885 @default.
- W3035331128 cites W2139090905 @default.
- W3035331128 cites W2152263452 @default.
- W3035331128 cites W2153579005 @default.
- W3035331128 cites W2154368244 @default.
- W3035331128 cites W2163605009 @default.
- W3035331128 cites W2170716495 @default.
- W3035331128 cites W2186780112 @default.
- W3035331128 cites W2251044566 @default.
- W3035331128 cites W2308720496 @default.
- W3035331128 cites W2620290674 @default.
- W3035331128 cites W2741986357 @default.
- W3035331128 cites W2798663534 @default.
- W3035331128 cites W2810565934 @default.
- W3035331128 cites W2950527759 @default.
- W3035331128 cites W2962739339 @default.
- W3035331128 cites W2962968135 @default.
- W3035331128 cites W2963073938 @default.
- W3035331128 cites W2963126845 @default.
- W3035331128 cites W2963267799 @default.
- W3035331128 cites W2963399829 @default.
- W3035331128 cites W2963403868 @default.
- W3035331128 cites W2963622428 @default.
- W3035331128 cites W2963655793 @default.
- W3035331128 cites W2963748792 @default.
- W3035331128 cites W2964121744 @default.
- W3035331128 cites W2964271186 @default.
- W3035331128 cites W2971252690 @default.
- W3035331128 cites W2971296908 @default.
- W3035331128 cites W2978165923 @default.
- W3035331128 cites W2979401726 @default.
- W3035331128 cites W3021713638 @default.
- W3035331128 cites W2963543962 @default.
- W3035331128 doi "https://doi.org/10.18653/v1/2020.acl-main.676" @default.
- W3035331128 hasPublicationYear "2020" @default.
- W3035331128 type Work @default.
- W3035331128 sameAs 3035331128 @default.
- W3035331128 citedByCount "117" @default.
- W3035331128 countsByYear W30353311282019 @default.
- W3035331128 countsByYear W30353311282020 @default.
- W3035331128 countsByYear W30353311282021 @default.
- W3035331128 countsByYear W30353311282022 @default.
- W3035331128 countsByYear W30353311282023 @default.
- W3035331128 crossrefType "proceedings-article" @default.
- W3035331128 hasAuthorship W3035331128A5070829652 @default.
- W3035331128 hasBestOaLocation W30353311281 @default.
- W3035331128 hasConcept C100279451 @default.
- W3035331128 hasConcept C111472728 @default.
- W3035331128 hasConcept C119857082 @default.
- W3035331128 hasConcept C136197465 @default.
- W3035331128 hasConcept C137293760 @default.
- W3035331128 hasConcept C138885662 @default.
- W3035331128 hasConcept C142724271 @default.
- W3035331128 hasConcept C154945302 @default.
- W3035331128 hasConcept C162324750 @default.
- W3035331128 hasConcept C186644900 @default.
- W3035331128 hasConcept C187736073 @default.
- W3035331128 hasConcept C197352929 @default.
- W3035331128 hasConcept C204321447 @default.
- W3035331128 hasConcept C204787440 @default.
- W3035331128 hasConcept C2778112365 @default.
- W3035331128 hasConcept C2780385302 @default.
- W3035331128 hasConcept C2780451532 @default.
- W3035331128 hasConcept C2780586882 @default.
- W3035331128 hasConcept C28006648 @default.
- W3035331128 hasConcept C35639132 @default.
- W3035331128 hasConcept C40969351 @default.
- W3035331128 hasConcept C41008148 @default.
- W3035331128 hasConcept C51632099 @default.
- W3035331128 hasConcept C54355233 @default.
- W3035331128 hasConcept C71924100 @default.
- W3035331128 hasConcept C86803240 @default.
- W3035331128 hasConceptScore W3035331128C100279451 @default.
- W3035331128 hasConceptScore W3035331128C111472728 @default.
- W3035331128 hasConceptScore W3035331128C119857082 @default.
- W3035331128 hasConceptScore W3035331128C136197465 @default.
- W3035331128 hasConceptScore W3035331128C137293760 @default.
- W3035331128 hasConceptScore W3035331128C138885662 @default.
- W3035331128 hasConceptScore W3035331128C142724271 @default.
- W3035331128 hasConceptScore W3035331128C154945302 @default.
- W3035331128 hasConceptScore W3035331128C162324750 @default.
- W3035331128 hasConceptScore W3035331128C186644900 @default.
- W3035331128 hasConceptScore W3035331128C187736073 @default.