Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366990295> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4366990295 abstract "In this paper, we propose a robustness benchmark for image-text matching models to assess their vulnerabilities. To this end, we insert adversarial texts and images into the search pool (i.e., gallery set) and evaluate models with the adversarial data. Specifically, we replace a word in the text to change the meaning of the text and mix images with different images to create perceptible changes in pixels. We assume that such explicit alterations would not deceive a robust model, as they should understand the holistic meaning of texts and images simultaneously. However, in our evaluations on the proposed benchmark, many state-of-the-art models show significant performance degradation, e.g., Recall@1: 81.9% $rightarrow$ 64.5% in BLIP, 66.1% $rightarrow$ 37.5% in VSE$infty$, where the models favor adversarial texts/images over the original ones. This reveals the current vision-language models may not account for subtle changes or understand the overall context of texts and images. Our findings can provide insights for improving the robustness of the vision-language models and devising more diverse stress-test methods in cross-modal retrieval task. Source code and dataset will be available at https://github.com/pseulki/rococo." @default.
- W4366990295 created "2023-04-27" @default.
- W4366990295 creator A5003220288 @default.
- W4366990295 creator A5012065589 @default.
- W4366990295 creator A5018158717 @default.
- W4366990295 creator A5029568033 @default.
- W4366990295 creator A5067167236 @default.
- W4366990295 creator A5078915948 @default.
- W4366990295 date "2023-04-20" @default.
- W4366990295 modified "2023-09-24" @default.
- W4366990295 title "RoCOCO: Robustness Benchmark of MS-COCO to Stress-test Image-Text Matching Models" @default.
- W4366990295 doi "https://doi.org/10.48550/arxiv.2304.10727" @default.
- W4366990295 hasPublicationYear "2023" @default.
- W4366990295 type Work @default.
- W4366990295 citedByCount "0" @default.
- W4366990295 crossrefType "posted-content" @default.
- W4366990295 hasAuthorship W4366990295A5003220288 @default.
- W4366990295 hasAuthorship W4366990295A5012065589 @default.
- W4366990295 hasAuthorship W4366990295A5018158717 @default.
- W4366990295 hasAuthorship W4366990295A5029568033 @default.
- W4366990295 hasAuthorship W4366990295A5067167236 @default.
- W4366990295 hasAuthorship W4366990295A5078915948 @default.
- W4366990295 hasBestOaLocation W43669902951 @default.
- W4366990295 hasConcept C104317684 @default.
- W4366990295 hasConcept C119857082 @default.
- W4366990295 hasConcept C13280743 @default.
- W4366990295 hasConcept C137293760 @default.
- W4366990295 hasConcept C154945302 @default.
- W4366990295 hasConcept C160633673 @default.
- W4366990295 hasConcept C169903167 @default.
- W4366990295 hasConcept C177264268 @default.
- W4366990295 hasConcept C185592680 @default.
- W4366990295 hasConcept C185798385 @default.
- W4366990295 hasConcept C199360897 @default.
- W4366990295 hasConcept C204321447 @default.
- W4366990295 hasConcept C205649164 @default.
- W4366990295 hasConcept C2776760102 @default.
- W4366990295 hasConcept C37736160 @default.
- W4366990295 hasConcept C41008148 @default.
- W4366990295 hasConcept C55493867 @default.
- W4366990295 hasConcept C63479239 @default.
- W4366990295 hasConceptScore W4366990295C104317684 @default.
- W4366990295 hasConceptScore W4366990295C119857082 @default.
- W4366990295 hasConceptScore W4366990295C13280743 @default.
- W4366990295 hasConceptScore W4366990295C137293760 @default.
- W4366990295 hasConceptScore W4366990295C154945302 @default.
- W4366990295 hasConceptScore W4366990295C160633673 @default.
- W4366990295 hasConceptScore W4366990295C169903167 @default.
- W4366990295 hasConceptScore W4366990295C177264268 @default.
- W4366990295 hasConceptScore W4366990295C185592680 @default.
- W4366990295 hasConceptScore W4366990295C185798385 @default.
- W4366990295 hasConceptScore W4366990295C199360897 @default.
- W4366990295 hasConceptScore W4366990295C204321447 @default.
- W4366990295 hasConceptScore W4366990295C205649164 @default.
- W4366990295 hasConceptScore W4366990295C2776760102 @default.
- W4366990295 hasConceptScore W4366990295C37736160 @default.
- W4366990295 hasConceptScore W4366990295C41008148 @default.
- W4366990295 hasConceptScore W4366990295C55493867 @default.
- W4366990295 hasConceptScore W4366990295C63479239 @default.
- W4366990295 hasLocation W43669902951 @default.
- W4366990295 hasLocation W43669902952 @default.
- W4366990295 hasOpenAccess W4366990295 @default.
- W4366990295 hasPrimaryLocation W43669902951 @default.
- W4366990295 hasRelatedWork W1772447446 @default.
- W4366990295 hasRelatedWork W2943368551 @default.
- W4366990295 hasRelatedWork W3035729345 @default.
- W4366990295 hasRelatedWork W3177252557 @default.
- W4366990295 hasRelatedWork W4288362200 @default.
- W4366990295 hasRelatedWork W4311734044 @default.
- W4366990295 hasRelatedWork W4376606861 @default.
- W4366990295 hasRelatedWork W4378945634 @default.
- W4366990295 hasRelatedWork W4379255972 @default.
- W4366990295 hasRelatedWork W4383955378 @default.
- W4366990295 isParatext "false" @default.
- W4366990295 isRetracted "false" @default.
- W4366990295 workType "article" @default.