FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313476629> ?p ?o ?g. }

• W4313476629 endingPage "31" @default.
• W4313476629 startingPage "1" @default.
• W4313476629 abstract "Neural document ranking models perform impressively well due to superior language understanding gained from pre-training tasks. However, due to their complexity and large number of parameters these (typically transformer-based) models are often non-interpretable in that ranking decisions can not be clearly attributed to specific parts of the input documents. In this article, we propose ranking models that are inherently interpretable by generating explanations as a by-product of the prediction decision. We introduce the Select-And-Rank paradigm for document ranking, where we first output an explanation as a selected subset of sentences in a document. Thereafter, we solely use the explanation or selection to make the prediction, making explanations first-class citizens in the ranking process. Technically, we treat sentence selection as a latent variable trained jointly with the ranker from the final output. To that end, we propose an end-to-end training technique for Select-And-Rank models utilizing reparameterizable subset sampling using the Gumbel-max trick . We conduct extensive experiments to demonstrate that our approach is competitive to state-of-the-art methods. Our approach is broadly applicable to numerous ranking tasks and furthers the goal of building models that are interpretable by design . Finally, we present real-world applications that benefit from our sentence selection method." @default.
• W4313476629 created "2023-01-06" @default.
• W4313476629 creator A5064040097 @default.
• W4313476629 creator A5071792143 @default.
• W4313476629 creator A5075681290 @default.
• W4313476629 date "2023-03-23" @default.
• W4313476629 modified "2023-10-17" @default.
• W4313476629 title "Extractive Explanations for Interpretable Text Ranking" @default.
• W4313476629 cites W2131876387 @default.
• W4313476629 cites W2136189984 @default.
• W4313476629 cites W2186845332 @default.
• W4313476629 cites W2267186426 @default.
• W4313476629 cites W2286300105 @default.
• W4313476629 cites W2470673105 @default.
• W4313476629 cites W2517782820 @default.
• W4313476629 cites W2534432321 @default.
• W4313476629 cites W2536015822 @default.
• W4313476629 cites W2539671052 @default.
• W4313476629 cites W2648699835 @default.
• W4313476629 cites W2735155365 @default.
• W4313476629 cites W2782157559 @default.
• W4313476629 cites W2783640434 @default.
• W4313476629 cites W2798631888 @default.
• W4313476629 cites W2890294942 @default.
• W4313476629 cites W2890881127 @default.
• W4313476629 cites W2891482011 @default.
• W4313476629 cites W2934842096 @default.
• W4313476629 cites W2945127593 @default.
• W4313476629 cites W2945976633 @default.
• W4313476629 cites W2949227999 @default.
• W4313476629 cites W2955013659 @default.
• W4313476629 cites W2963029083 @default.
• W4313476629 cites W2963233086 @default.
• W4313476629 cites W2963961878 @default.
• W4313476629 cites W2964012472 @default.
• W4313476629 cites W2964554227 @default.
• W4313476629 cites W2970726176 @default.
• W4313476629 cites W2971209824 @default.
• W4313476629 cites W3012594078 @default.
• W4313476629 cites W3013870856 @default.
• W4313476629 cites W3021397474 @default.
• W4313476629 cites W3034917890 @default.
• W4313476629 cites W3035503910 @default.
• W4313476629 cites W3044284384 @default.
• W4313476629 cites W3093655911 @default.
• W4313476629 cites W3093860123 @default.
• W4313476629 cites W3097252660 @default.
• W4313476629 cites W3099990142 @default.
• W4313476629 cites W3102286003 @default.
• W4313476629 cites W3103101828 @default.
• W4313476629 cites W3104486441 @default.
• W4313476629 cites W3105095409 @default.
• W4313476629 cites W3112008364 @default.
• W4313476629 cites W3113553002 @default.
• W4313476629 cites W3123856867 @default.
• W4313476629 cites W3153184434 @default.
• W4313476629 cites W3155114168 @default.
• W4313476629 cites W3156413894 @default.
• W4313476629 cites W3161851957 @default.
• W4313476629 cites W3168461525 @default.
• W4313476629 cites W3170572542 @default.
• W4313476629 cites W3184918446 @default.
• W4313476629 cites W3197223198 @default.
• W4313476629 cites W3197577391 @default.
• W4313476629 cites W3211142893 @default.
• W4313476629 cites W4224308764 @default.
• W4313476629 cites W4226143364 @default.
• W4313476629 cites W4229078373 @default.
• W4313476629 cites W4246858749 @default.
• W4313476629 cites W4252076394 @default.
• W4313476629 cites W4285192297 @default.
• W4313476629 cites W4288086191 @default.
• W4313476629 cites W83993254 @default.
• W4313476629 doi "https://doi.org/10.1145/3576924" @default.
• W4313476629 hasPublicationYear "2023" @default.
• W4313476629 type Work @default.
• W4313476629 citedByCount "2" @default.
• W4313476629 countsByYear W43134766292023 @default.
• W4313476629 crossrefType "journal-article" @default.
• W4313476629 hasAuthorship W4313476629A5064040097 @default.
• W4313476629 hasAuthorship W4313476629A5071792143 @default.
• W4313476629 hasAuthorship W4313476629A5075681290 @default.
• W4313476629 hasBestOaLocation W43134766291 @default.
• W4313476629 hasConcept C114614502 @default.
• W4313476629 hasConcept C119857082 @default.
• W4313476629 hasConcept C121332964 @default.
• W4313476629 hasConcept C154945302 @default.
• W4313476629 hasConcept C164226766 @default.
• W4313476629 hasConcept C165801399 @default.
• W4313476629 hasConcept C189430467 @default.
• W4313476629 hasConcept C204321447 @default.
• W4313476629 hasConcept C2777530160 @default.
• W4313476629 hasConcept C33923547 @default.
• W4313476629 hasConcept C41008148 @default.
• W4313476629 hasConcept C62520636 @default.
• W4313476629 hasConcept C66322947 @default.
• W4313476629 hasConcept C81917197 @default.
• W4313476629 hasConcept C86037889 @default.

• next