Matches in SemOpenAlex for { <https://semopenalex.org/work/W3215769852> ?p ?o ?g. }
- W3215769852 abstract "Deep learning has recently made remarkable progress in natural language processing. Yet, the resulting algorithms remain far from competing with the language abilities of the human brain. Predictive coding theory offers a potential explanation to this discrepancy: while deep language algorithms are optimized to predict adjacent words, the human brain would be tuned to make long-range and hierarchical predictions. To test this hypothesis, we analyze the fMRI brain signals of 304 subjects each listening to 70min of short stories. After confirming that the activations of deep language algorithms linearly map onto those of the brain, we show that enhancing these models with long-range forecast representations improves their brain-mapping. The results further reveal a hierarchy of predictions in the brain, whereby the fronto-parietal cortices forecast more abstract and more distant representations than the temporal cortices. Overall, this study strengthens predictive coding theory and suggests a critical role of long-range and hierarchical predictions in natural language processing." @default.
- W3215769852 created "2021-12-06" @default.
- W3215769852 creator A5018256474 @default.
- W3215769852 creator A5024428666 @default.
- W3215769852 creator A5056821038 @default.
- W3215769852 date "2021-11-28" @default.
- W3215769852 modified "2023-09-27" @default.
- W3215769852 title "Long-range and hierarchical language predictions in brains and algorithms" @default.
- W3215769852 cites W1579929927 @default.
- W3215769852 cites W1715013381 @default.
- W3215769852 cites W1969221307 @default.
- W3215769852 cites W1991825895 @default.
- W3215769852 cites W2040036684 @default.
- W3215769852 cites W2058616551 @default.
- W3215769852 cites W2073257493 @default.
- W3215769852 cites W2096375888 @default.
- W3215769852 cites W2097117768 @default.
- W3215769852 cites W2101234009 @default.
- W3215769852 cites W2119885245 @default.
- W3215769852 cites W2138210540 @default.
- W3215769852 cites W2170167574 @default.
- W3215769852 cites W2195506630 @default.
- W3215769852 cites W2344975321 @default.
- W3215769852 cites W2537084945 @default.
- W3215769852 cites W2606669569 @default.
- W3215769852 cites W2606837722 @default.
- W3215769852 cites W2616551000 @default.
- W3215769852 cites W2798664956 @default.
- W3215769852 cites W2805003518 @default.
- W3215769852 cites W2911109671 @default.
- W3215769852 cites W2922523190 @default.
- W3215769852 cites W2938704169 @default.
- W3215769852 cites W2947012833 @default.
- W3215769852 cites W2949417144 @default.
- W3215769852 cites W2951583631 @default.
- W3215769852 cites W2963341956 @default.
- W3215769852 cites W2963403868 @default.
- W3215769852 cites W2965373594 @default.
- W3215769852 cites W2970597249 @default.
- W3215769852 cites W2977009803 @default.
- W3215769852 cites W2987283559 @default.
- W3215769852 cites W2991642412 @default.
- W3215769852 cites W2997938465 @default.
- W3215769852 cites W2997968655 @default.
- W3215769852 cites W3005680577 @default.
- W3215769852 cites W3013571468 @default.
- W3215769852 cites W3030163527 @default.
- W3215769852 cites W3037273551 @default.
- W3215769852 cites W3039556919 @default.
- W3215769852 cites W3047499204 @default.
- W3215769852 cites W3082274269 @default.
- W3215769852 cites W3091562615 @default.
- W3215769852 cites W3092973197 @default.
- W3215769852 cites W3105728642 @default.
- W3215769852 cites W3108672584 @default.
- W3215769852 cites W3116202855 @default.
- W3215769852 cites W3118674731 @default.
- W3215769852 cites W3131305086 @default.
- W3215769852 cites W3133934889 @default.
- W3215769852 cites W3133995764 @default.
- W3215769852 cites W3134521268 @default.
- W3215769852 cites W3154773080 @default.
- W3215769852 cites W3155211769 @default.
- W3215769852 cites W3163602117 @default.
- W3215769852 cites W3171206729 @default.
- W3215769852 cites W3171589458 @default.
- W3215769852 cites W3206626273 @default.
- W3215769852 cites W3213177139 @default.
- W3215769852 cites W3213402928 @default.
- W3215769852 hasPublicationYear "2021" @default.
- W3215769852 type Work @default.
- W3215769852 sameAs 3215769852 @default.
- W3215769852 citedByCount "0" @default.
- W3215769852 crossrefType "posted-content" @default.
- W3215769852 hasAuthorship W3215769852A5018256474 @default.
- W3215769852 hasAuthorship W3215769852A5024428666 @default.
- W3215769852 hasAuthorship W3215769852A5056821038 @default.
- W3215769852 hasConcept C105795698 @default.
- W3215769852 hasConcept C137293760 @default.
- W3215769852 hasConcept C154945302 @default.
- W3215769852 hasConcept C15744967 @default.
- W3215769852 hasConcept C159985019 @default.
- W3215769852 hasConcept C162324750 @default.
- W3215769852 hasConcept C169760540 @default.
- W3215769852 hasConcept C177291462 @default.
- W3215769852 hasConcept C179518139 @default.
- W3215769852 hasConcept C188147891 @default.
- W3215769852 hasConcept C192562407 @default.
- W3215769852 hasConcept C195324797 @default.
- W3215769852 hasConcept C204321447 @default.
- W3215769852 hasConcept C204323151 @default.
- W3215769852 hasConcept C2777670902 @default.
- W3215769852 hasConcept C2778061373 @default.
- W3215769852 hasConcept C2983448237 @default.
- W3215769852 hasConcept C31170391 @default.
- W3215769852 hasConcept C33923547 @default.
- W3215769852 hasConcept C34447519 @default.
- W3215769852 hasConcept C41008148 @default.
- W3215769852 hasConcept C46312422 @default.
- W3215769852 hasConceptScore W3215769852C105795698 @default.