Matches in SemOpenAlex for { <https://semopenalex.org/work/W4229069940> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W4229069940 abstract "In recent years, the pre-training-then-fine-tuning paradigm has yielded immense success on a wide spectrum of cross-modal tasks, such as visual question answering (VQA), in which a visual-language (VL) model is first optimized via self-supervised task objectives, e.g., masked language modeling (MLM) and image-text matching (ITM), and then fine-tuned to adapt to downstream task (e.g., VQA) via a brand-new objective function, e.g., answer prediction. However, the inconsistency of the objective forms not only severely limits the generalization of pre-trained VL models to downstream tasks, but also requires a large amount of labeled data for fine-tuning. To alleviate the problem, we propose an innovative VL fine-tuning paradigm (named Declaration-based Prompt Tuning, abbreviated as DPT), which fine-tunes the model for downstream VQA using the pre-training objectives, boosting the effective adaptation of pre-trained models to the downstream task. Specifically, DPT reformulates the VQA task via (1) textual adaptation, which converts the given questions into declarative sentence form for prompt-tuning, and (2) task adaptation, which optimizes the objective function of VQA problem in the manner of pre-training phase. Experimental results on GQA dataset show that DPT outperforms the fine-tuned counterpart by a large margin regarding accuracy in both fully-supervised (2.68%) and zero-shot/fewshot (over 31%) settings. All the data and codes will be available to facilitate future research." @default.
- W4229069940 created "2022-05-08" @default.
- W4229069940 creator A5003692270 @default.
- W4229069940 creator A5003799076 @default.
- W4229069940 creator A5064684366 @default.
- W4229069940 creator A5066682332 @default.
- W4229069940 date "2022-07-01" @default.
- W4229069940 modified "2023-10-17" @default.
- W4229069940 title "Declaration-based Prompt Tuning for Visual Question Answering" @default.
- W4229069940 doi "https://doi.org/10.24963/ijcai.2022/450" @default.
- W4229069940 hasPublicationYear "2022" @default.
- W4229069940 type Work @default.
- W4229069940 citedByCount "0" @default.
- W4229069940 crossrefType "proceedings-article" @default.
- W4229069940 hasAuthorship W4229069940A5003692270 @default.
- W4229069940 hasAuthorship W4229069940A5003799076 @default.
- W4229069940 hasAuthorship W4229069940A5064684366 @default.
- W4229069940 hasAuthorship W4229069940A5066682332 @default.
- W4229069940 hasBestOaLocation W42290699401 @default.
- W4229069940 hasConcept C119857082 @default.
- W4229069940 hasConcept C120665830 @default.
- W4229069940 hasConcept C121332964 @default.
- W4229069940 hasConcept C137293760 @default.
- W4229069940 hasConcept C138147947 @default.
- W4229069940 hasConcept C139807058 @default.
- W4229069940 hasConcept C153083717 @default.
- W4229069940 hasConcept C154945302 @default.
- W4229069940 hasConcept C162324750 @default.
- W4229069940 hasConcept C187736073 @default.
- W4229069940 hasConcept C199360897 @default.
- W4229069940 hasConcept C204321447 @default.
- W4229069940 hasConcept C2777530160 @default.
- W4229069940 hasConcept C2780451532 @default.
- W4229069940 hasConcept C41008148 @default.
- W4229069940 hasConcept C44291984 @default.
- W4229069940 hasConcept C774472 @default.
- W4229069940 hasConceptScore W4229069940C119857082 @default.
- W4229069940 hasConceptScore W4229069940C120665830 @default.
- W4229069940 hasConceptScore W4229069940C121332964 @default.
- W4229069940 hasConceptScore W4229069940C137293760 @default.
- W4229069940 hasConceptScore W4229069940C138147947 @default.
- W4229069940 hasConceptScore W4229069940C139807058 @default.
- W4229069940 hasConceptScore W4229069940C153083717 @default.
- W4229069940 hasConceptScore W4229069940C154945302 @default.
- W4229069940 hasConceptScore W4229069940C162324750 @default.
- W4229069940 hasConceptScore W4229069940C187736073 @default.
- W4229069940 hasConceptScore W4229069940C199360897 @default.
- W4229069940 hasConceptScore W4229069940C204321447 @default.
- W4229069940 hasConceptScore W4229069940C2777530160 @default.
- W4229069940 hasConceptScore W4229069940C2780451532 @default.
- W4229069940 hasConceptScore W4229069940C41008148 @default.
- W4229069940 hasConceptScore W4229069940C44291984 @default.
- W4229069940 hasConceptScore W4229069940C774472 @default.
- W4229069940 hasLocation W42290699401 @default.
- W4229069940 hasLocation W42290699402 @default.
- W4229069940 hasOpenAccess W4229069940 @default.
- W4229069940 hasPrimaryLocation W42290699401 @default.
- W4229069940 hasRelatedWork W1517743118 @default.
- W4229069940 hasRelatedWork W20999564 @default.
- W4229069940 hasRelatedWork W2107004888 @default.
- W4229069940 hasRelatedWork W2747680751 @default.
- W4229069940 hasRelatedWork W3016473712 @default.
- W4229069940 hasRelatedWork W3107474891 @default.
- W4229069940 hasRelatedWork W3114537677 @default.
- W4229069940 hasRelatedWork W3194620391 @default.
- W4229069940 hasRelatedWork W4309086292 @default.
- W4229069940 hasRelatedWork W2972820682 @default.
- W4229069940 isParatext "false" @default.
- W4229069940 isRetracted "false" @default.
- W4229069940 workType "article" @default.