Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313218048> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4313218048 abstract "Despite the impressive improvements of Visual Question Answer (VQA), it still remains a challenge of how to avoid the suffering of spurious correlations from textual content to answer. Previous researches have shown that due to the existence of language bias in the VQA dataset, VQA models may tend to capture superficial statistical correlation and suffer from the poor generalization capability in the out-of-distribution data. To alleviate the biases caused by language modality, we propose a method of context augmentation and adaptive loss adjustment, which can alleviate shortcut learning behavior of VQA models. Specifically, the existence of language bias is due to the high co-occurrence frequency of categories and the words in “Question”, therefore, we propose to use “Paraphrase Generation” to produce paraphrases with diverse contexts, so as to mitigate such correlation. Secondly, we use adaptive loss adjustment to adjust the importance of samples, that is, reduce the importance of bias-aligned samples and improve the importance of bias-conflicting samples, so as to guide the model to capture the intrinsic attributes that are beneficial to generalization. The experiments have demonstrated the feasibility and validity of our method on a variety of VQA models." @default.
- W4313218048 created "2023-01-06" @default.
- W4313218048 creator A5009988205 @default.
- W4313218048 creator A5023135578 @default.
- W4313218048 creator A5081360719 @default.
- W4313218048 date "2022-12-29" @default.
- W4313218048 modified "2023-09-27" @default.
- W4313218048 title "Alleviating shortcut learning behavior of VQA model with context augmentation and adaptive loss adjustment" @default.
- W4313218048 doi "https://doi.org/10.1117/12.2661996" @default.
- W4313218048 hasPublicationYear "2022" @default.
- W4313218048 type Work @default.
- W4313218048 citedByCount "0" @default.
- W4313218048 crossrefType "proceedings-article" @default.
- W4313218048 hasAuthorship W4313218048A5009988205 @default.
- W4313218048 hasAuthorship W4313218048A5023135578 @default.
- W4313218048 hasAuthorship W4313218048A5081360719 @default.
- W4313218048 hasBestOaLocation W43132180481 @default.
- W4313218048 hasConcept C117220453 @default.
- W4313218048 hasConcept C119857082 @default.
- W4313218048 hasConcept C134306372 @default.
- W4313218048 hasConcept C136197465 @default.
- W4313218048 hasConcept C137293760 @default.
- W4313218048 hasConcept C149782125 @default.
- W4313218048 hasConcept C151730666 @default.
- W4313218048 hasConcept C154945302 @default.
- W4313218048 hasConcept C177148314 @default.
- W4313218048 hasConcept C204321447 @default.
- W4313218048 hasConcept C207609745 @default.
- W4313218048 hasConcept C2524010 @default.
- W4313218048 hasConcept C2779343474 @default.
- W4313218048 hasConcept C2780922921 @default.
- W4313218048 hasConcept C33923547 @default.
- W4313218048 hasConcept C41008148 @default.
- W4313218048 hasConcept C86803240 @default.
- W4313218048 hasConcept C97256817 @default.
- W4313218048 hasConceptScore W4313218048C117220453 @default.
- W4313218048 hasConceptScore W4313218048C119857082 @default.
- W4313218048 hasConceptScore W4313218048C134306372 @default.
- W4313218048 hasConceptScore W4313218048C136197465 @default.
- W4313218048 hasConceptScore W4313218048C137293760 @default.
- W4313218048 hasConceptScore W4313218048C149782125 @default.
- W4313218048 hasConceptScore W4313218048C151730666 @default.
- W4313218048 hasConceptScore W4313218048C154945302 @default.
- W4313218048 hasConceptScore W4313218048C177148314 @default.
- W4313218048 hasConceptScore W4313218048C204321447 @default.
- W4313218048 hasConceptScore W4313218048C207609745 @default.
- W4313218048 hasConceptScore W4313218048C2524010 @default.
- W4313218048 hasConceptScore W4313218048C2779343474 @default.
- W4313218048 hasConceptScore W4313218048C2780922921 @default.
- W4313218048 hasConceptScore W4313218048C33923547 @default.
- W4313218048 hasConceptScore W4313218048C41008148 @default.
- W4313218048 hasConceptScore W4313218048C86803240 @default.
- W4313218048 hasConceptScore W4313218048C97256817 @default.
- W4313218048 hasLocation W43132180481 @default.
- W4313218048 hasOpenAccess W4313218048 @default.
- W4313218048 hasPrimaryLocation W43132180481 @default.
- W4313218048 hasRelatedWork W2003010043 @default.
- W4313218048 hasRelatedWork W2135591339 @default.
- W4313218048 hasRelatedWork W2145637785 @default.
- W4313218048 hasRelatedWork W2359001871 @default.
- W4313218048 hasRelatedWork W2398825887 @default.
- W4313218048 hasRelatedWork W2943490406 @default.
- W4313218048 hasRelatedWork W3203726166 @default.
- W4313218048 hasRelatedWork W3212368093 @default.
- W4313218048 hasRelatedWork W4320467357 @default.
- W4313218048 hasRelatedWork W4385574294 @default.
- W4313218048 isParatext "false" @default.
- W4313218048 isRetracted "false" @default.
- W4313218048 workType "article" @default.