Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386158758> ?p ?o ?g. }
Showing items 1 to 64 of
64
with 100 items per page.
- W4386158758 abstract "Prompt tuning, like CoOp, has recently shown promising vision recognizing and transfer learning ability on various downstream tasks with the emergence of large pre-trained vision-language models like CLIP. However, we identify that existing uni-modal prompt tuning approaches may result in sub-optimal performance since this uni-modal design breaks the original alignment of textual and visual representations in the pre-trained model. Inspired by the nature of pre-trained vision-language models, we aim to achieve completeness in prompt tuning and propose a novel approach called Multi-modal Deep-symphysis Prompt Tuning, dubbed as MuDPT, which extends independent multi-modal prompt tuning by additionally learning a model-agnostic transformative network to allow deep hierarchical bi-directional prompt fusion. We evaluate the effectiveness of MuDPT on few-shot vision recognition and out-of-domain generalization tasks. Compared with the state-of-the-art methods, MuDPT achieves better recognition and generalization ability with an apparent margin thanks to synergistic alignment of textual and visual representations. Our code is available at: https://github.com/Mechrev0/MuDPT." @default.
- W4386158758 created "2023-08-26" @default.
- W4386158758 creator A5012889189 @default.
- W4386158758 creator A5035752387 @default.
- W4386158758 creator A5042236737 @default.
- W4386158758 creator A5068492996 @default.
- W4386158758 date "2023-07-01" @default.
- W4386158758 modified "2023-10-16" @default.
- W4386158758 title "MuDPT: Multi-modal Deep-symphysis Prompt Tuning for Large Pre-trained Vision-Language Models" @default.
- W4386158758 cites W2108598243 @default.
- W4386158758 cites W3198377975 @default.
- W4386158758 cites W4226058394 @default.
- W4386158758 cites W4285191490 @default.
- W4386158758 cites W4312310776 @default.
- W4386158758 doi "https://doi.org/10.1109/icme55011.2023.00013" @default.
- W4386158758 hasPublicationYear "2023" @default.
- W4386158758 type Work @default.
- W4386158758 citedByCount "0" @default.
- W4386158758 crossrefType "proceedings-article" @default.
- W4386158758 hasAuthorship W4386158758A5012889189 @default.
- W4386158758 hasAuthorship W4386158758A5035752387 @default.
- W4386158758 hasAuthorship W4386158758A5042236737 @default.
- W4386158758 hasAuthorship W4386158758A5068492996 @default.
- W4386158758 hasConcept C108583219 @default.
- W4386158758 hasConcept C119857082 @default.
- W4386158758 hasConcept C134306372 @default.
- W4386158758 hasConcept C154945302 @default.
- W4386158758 hasConcept C177148314 @default.
- W4386158758 hasConcept C185592680 @default.
- W4386158758 hasConcept C188027245 @default.
- W4386158758 hasConcept C2983448237 @default.
- W4386158758 hasConcept C33923547 @default.
- W4386158758 hasConcept C41008148 @default.
- W4386158758 hasConcept C71139939 @default.
- W4386158758 hasConcept C774472 @default.
- W4386158758 hasConceptScore W4386158758C108583219 @default.
- W4386158758 hasConceptScore W4386158758C119857082 @default.
- W4386158758 hasConceptScore W4386158758C134306372 @default.
- W4386158758 hasConceptScore W4386158758C154945302 @default.
- W4386158758 hasConceptScore W4386158758C177148314 @default.
- W4386158758 hasConceptScore W4386158758C185592680 @default.
- W4386158758 hasConceptScore W4386158758C188027245 @default.
- W4386158758 hasConceptScore W4386158758C2983448237 @default.
- W4386158758 hasConceptScore W4386158758C33923547 @default.
- W4386158758 hasConceptScore W4386158758C41008148 @default.
- W4386158758 hasConceptScore W4386158758C71139939 @default.
- W4386158758 hasConceptScore W4386158758C774472 @default.
- W4386158758 hasFunder F4320337504 @default.
- W4386158758 hasLocation W43861587581 @default.
- W4386158758 hasOpenAccess W4386158758 @default.
- W4386158758 hasPrimaryLocation W43861587581 @default.
- W4386158758 hasRelatedWork W3014300295 @default.
- W4386158758 hasRelatedWork W3164822677 @default.
- W4386158758 hasRelatedWork W4223943233 @default.
- W4386158758 hasRelatedWork W4225161397 @default.
- W4386158758 hasRelatedWork W4250304930 @default.
- W4386158758 hasRelatedWork W4312200629 @default.
- W4386158758 hasRelatedWork W4360585206 @default.
- W4386158758 hasRelatedWork W4364306694 @default.
- W4386158758 hasRelatedWork W4380075502 @default.
- W4386158758 hasRelatedWork W4380086463 @default.
- W4386158758 isParatext "false" @default.
- W4386158758 isRetracted "false" @default.
- W4386158758 workType "article" @default.