Matches in SemOpenAlex for { <https://semopenalex.org/work/W3200919998> ?p ?o ?g. }
- W3200919998 abstract "Current pre-trained models applied to summarization are prone to factual inconsistencies which either misrepresent the source text or introduce extraneous information. Thus, comparing the factual consistency of summaries is necessary as we develop improved models. However, the optimal human evaluation setup for factual consistency has not been standardized. To address this issue, we crowdsourced evaluations for factual consistency using the rating-based Likert scale and ranking-based Best-Worst Scaling protocols, on 100 articles from each of the CNN-Daily Mail and XSum datasets over four state-of-the-art models, to determine the most reliable evaluation framework. We find that ranking-based protocols offer a more reliable measure of summary quality across datasets, while the reliability of Likert ratings depends on the target dataset and the evaluation design. Our crowdsourcing templates and summary evaluations will be publicly available to facilitate future research on factual consistency in summarization." @default.
- W3200919998 created "2021-09-27" @default.
- W3200919998 creator A5002207077 @default.
- W3200919998 creator A5006794114 @default.
- W3200919998 creator A5023648764 @default.
- W3200919998 creator A5034426796 @default.
- W3200919998 creator A5041681353 @default.
- W3200919998 creator A5067997502 @default.
- W3200919998 creator A5081787254 @default.
- W3200919998 creator A5082094976 @default.
- W3200919998 date "2021-09-19" @default.
- W3200919998 modified "2023-09-27" @default.
- W3200919998 title "Investigating Crowdsourcing Protocols for Evaluating the Factual Consistency of Summaries" @default.
- W3200919998 cites W136732505 @default.
- W3200919998 cites W1486839280 @default.
- W3200919998 cites W1544827683 @default.
- W3200919998 cites W1986551946 @default.
- W3200919998 cites W1988584482 @default.
- W3200919998 cites W2120481102 @default.
- W3200919998 cites W2170822781 @default.
- W3200919998 cites W2250332012 @default.
- W3200919998 cites W2783136253 @default.
- W3200919998 cites W2842624112 @default.
- W3200919998 cites W2888482885 @default.
- W3200919998 cites W2922709902 @default.
- W3200919998 cites W2944815030 @default.
- W3200919998 cites W2945760033 @default.
- W3200919998 cites W2951211142 @default.
- W3200919998 cites W2962785754 @default.
- W3200919998 cites W2962806234 @default.
- W3200919998 cites W2963047186 @default.
- W3200919998 cites W2963929190 @default.
- W3200919998 cites W2970419734 @default.
- W3200919998 cites W2970892365 @default.
- W3200919998 cites W2971274815 @default.
- W3200919998 cites W2982399380 @default.
- W3200919998 cites W2996614149 @default.
- W3200919998 cites W3026758008 @default.
- W3200919998 cites W3034188538 @default.
- W3200919998 cites W3034383590 @default.
- W3200919998 cites W3034715004 @default.
- W3200919998 cites W3087063344 @default.
- W3200919998 cites W3094244719 @default.
- W3200919998 cites W3099766584 @default.
- W3200919998 cites W3100439847 @default.
- W3200919998 cites W3102645206 @default.
- W3200919998 cites W3107229750 @default.
- W3200919998 cites W3153338149 @default.
- W3200919998 cites W3159259047 @default.
- W3200919998 cites W3170432046 @default.
- W3200919998 doi "https://doi.org/10.48550/arxiv.2109.09195" @default.
- W3200919998 hasPublicationYear "2021" @default.
- W3200919998 type Work @default.
- W3200919998 sameAs 3200919998 @default.
- W3200919998 citedByCount "0" @default.
- W3200919998 crossrefType "posted-content" @default.
- W3200919998 hasAuthorship W3200919998A5002207077 @default.
- W3200919998 hasAuthorship W3200919998A5006794114 @default.
- W3200919998 hasAuthorship W3200919998A5023648764 @default.
- W3200919998 hasAuthorship W3200919998A5034426796 @default.
- W3200919998 hasAuthorship W3200919998A5041681353 @default.
- W3200919998 hasAuthorship W3200919998A5067997502 @default.
- W3200919998 hasAuthorship W3200919998A5081787254 @default.
- W3200919998 hasAuthorship W3200919998A5082094976 @default.
- W3200919998 hasBestOaLocation W32009199981 @default.
- W3200919998 hasConcept C105776082 @default.
- W3200919998 hasConcept C105795698 @default.
- W3200919998 hasConcept C111472728 @default.
- W3200919998 hasConcept C121332964 @default.
- W3200919998 hasConcept C124101348 @default.
- W3200919998 hasConcept C136764020 @default.
- W3200919998 hasConcept C138885662 @default.
- W3200919998 hasConcept C154945302 @default.
- W3200919998 hasConcept C163258240 @default.
- W3200919998 hasConcept C170858558 @default.
- W3200919998 hasConcept C189430467 @default.
- W3200919998 hasConcept C23123220 @default.
- W3200919998 hasConcept C2522767166 @default.
- W3200919998 hasConcept C2776436953 @default.
- W3200919998 hasConcept C2779530757 @default.
- W3200919998 hasConcept C2780009758 @default.
- W3200919998 hasConcept C33923547 @default.
- W3200919998 hasConcept C41008148 @default.
- W3200919998 hasConcept C43214815 @default.
- W3200919998 hasConcept C62230096 @default.
- W3200919998 hasConcept C62520636 @default.
- W3200919998 hasConceptScore W3200919998C105776082 @default.
- W3200919998 hasConceptScore W3200919998C105795698 @default.
- W3200919998 hasConceptScore W3200919998C111472728 @default.
- W3200919998 hasConceptScore W3200919998C121332964 @default.
- W3200919998 hasConceptScore W3200919998C124101348 @default.
- W3200919998 hasConceptScore W3200919998C136764020 @default.
- W3200919998 hasConceptScore W3200919998C138885662 @default.
- W3200919998 hasConceptScore W3200919998C154945302 @default.
- W3200919998 hasConceptScore W3200919998C163258240 @default.
- W3200919998 hasConceptScore W3200919998C170858558 @default.
- W3200919998 hasConceptScore W3200919998C189430467 @default.
- W3200919998 hasConceptScore W3200919998C23123220 @default.
- W3200919998 hasConceptScore W3200919998C2522767166 @default.
- W3200919998 hasConceptScore W3200919998C2776436953 @default.