Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963206148> ?p ?o ?g. }
- W2963206148 abstract "Sequence-to-sequence neural network models for generation of conversational responses tend to generate safe, commonplace responses (e.g., I don’t know) regardless of the input. We suggest that the traditional objective function, i.e., the likelihood of output (response) given input (message) is unsuited to response generation tasks. Instead we propose using Maximum Mutual Information (MMI) as the objective function in neural models. Experimental results demonstrate that the proposed MMI models produce more diverse, interesting, and appropriate responses, yielding substantive gains in BLEU scores on two conversational datasets and in human evaluations." @default.
- W2963206148 created "2019-07-30" @default.
- W2963206148 creator A5015656536 @default.
- W2963206148 creator A5037066465 @default.
- W2963206148 creator A5047233371 @default.
- W2963206148 creator A5057693146 @default.
- W2963206148 creator A5077851706 @default.
- W2963206148 date "2016-01-01" @default.
- W2963206148 modified "2023-10-15" @default.
- W2963206148 title "A Diversity-Promoting Objective Function for Neural Conversation Models" @default.
- W2963206148 cites W10957333 @default.
- W2963206148 cites W1518951372 @default.
- W2963206148 cites W1519256446 @default.
- W2963206148 cites W1538748823 @default.
- W2963206148 cites W1578856370 @default.
- W2963206148 cites W1591706642 @default.
- W2963206148 cites W1604513301 @default.
- W2963206148 cites W1623245510 @default.
- W2963206148 cites W1847211030 @default.
- W2963206148 cites W1877570817 @default.
- W2963206148 cites W194577561 @default.
- W2963206148 cites W1948566616 @default.
- W2963206148 cites W1970207841 @default.
- W2963206148 cites W2004637830 @default.
- W2963206148 cites W2064675550 @default.
- W2963206148 cites W2083305840 @default.
- W2963206148 cites W2101105183 @default.
- W2963206148 cites W2115101920 @default.
- W2963206148 cites W2118434577 @default.
- W2963206148 cites W2124807415 @default.
- W2963206148 cites W2130119531 @default.
- W2963206148 cites W2130942839 @default.
- W2963206148 cites W2132997613 @default.
- W2963206148 cites W2140054881 @default.
- W2963206148 cites W2146574666 @default.
- W2963206148 cites W2250445771 @default.
- W2963206148 cites W2252065493 @default.
- W2963206148 cites W2296712013 @default.
- W2963206148 cites W2950178297 @default.
- W2963206148 cites W2962706528 @default.
- W2963206148 cites W2962883855 @default.
- W2963206148 cites W2963069010 @default.
- W2963206148 cites W2963963856 @default.
- W2963206148 cites W2964308564 @default.
- W2963206148 cites W319421170 @default.
- W2963206148 cites W635530177 @default.
- W2963206148 doi "https://doi.org/10.18653/v1/n16-1014" @default.
- W2963206148 hasPublicationYear "2016" @default.
- W2963206148 type Work @default.
- W2963206148 sameAs 2963206148 @default.
- W2963206148 citedByCount "1313" @default.
- W2963206148 countsByYear W29632061482015 @default.
- W2963206148 countsByYear W29632061482016 @default.
- W2963206148 countsByYear W29632061482017 @default.
- W2963206148 countsByYear W29632061482018 @default.
- W2963206148 countsByYear W29632061482019 @default.
- W2963206148 countsByYear W29632061482020 @default.
- W2963206148 countsByYear W29632061482021 @default.
- W2963206148 countsByYear W29632061482022 @default.
- W2963206148 countsByYear W29632061482023 @default.
- W2963206148 crossrefType "proceedings-article" @default.
- W2963206148 hasAuthorship W2963206148A5015656536 @default.
- W2963206148 hasAuthorship W2963206148A5037066465 @default.
- W2963206148 hasAuthorship W2963206148A5047233371 @default.
- W2963206148 hasAuthorship W2963206148A5057693146 @default.
- W2963206148 hasAuthorship W2963206148A5077851706 @default.
- W2963206148 hasBestOaLocation W29632061481 @default.
- W2963206148 hasConcept C119857082 @default.
- W2963206148 hasConcept C14036430 @default.
- W2963206148 hasConcept C144024400 @default.
- W2963206148 hasConcept C154945302 @default.
- W2963206148 hasConcept C15744967 @default.
- W2963206148 hasConcept C19165224 @default.
- W2963206148 hasConcept C204321447 @default.
- W2963206148 hasConcept C2777200299 @default.
- W2963206148 hasConcept C2778112365 @default.
- W2963206148 hasConcept C2781316041 @default.
- W2963206148 hasConcept C28490314 @default.
- W2963206148 hasConcept C41008148 @default.
- W2963206148 hasConcept C46312422 @default.
- W2963206148 hasConcept C50644808 @default.
- W2963206148 hasConcept C54355233 @default.
- W2963206148 hasConcept C78458016 @default.
- W2963206148 hasConcept C86803240 @default.
- W2963206148 hasConceptScore W2963206148C119857082 @default.
- W2963206148 hasConceptScore W2963206148C14036430 @default.
- W2963206148 hasConceptScore W2963206148C144024400 @default.
- W2963206148 hasConceptScore W2963206148C154945302 @default.
- W2963206148 hasConceptScore W2963206148C15744967 @default.
- W2963206148 hasConceptScore W2963206148C19165224 @default.
- W2963206148 hasConceptScore W2963206148C204321447 @default.
- W2963206148 hasConceptScore W2963206148C2777200299 @default.
- W2963206148 hasConceptScore W2963206148C2778112365 @default.
- W2963206148 hasConceptScore W2963206148C2781316041 @default.
- W2963206148 hasConceptScore W2963206148C28490314 @default.
- W2963206148 hasConceptScore W2963206148C41008148 @default.
- W2963206148 hasConceptScore W2963206148C46312422 @default.
- W2963206148 hasConceptScore W2963206148C50644808 @default.
- W2963206148 hasConceptScore W2963206148C54355233 @default.
- W2963206148 hasConceptScore W2963206148C78458016 @default.