Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384263460> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4384263460 abstract "Despite the recent progress in incremental learning, addressing catastrophic forgetting under distributional drift is still an open and important problem. Indeed, while state-of-the-art domain incremental learning (DIL) methods perform satisfactorily within known domains, their performance largely degrades in the presence of novel domains. This limitation hampers their generalizability, and restricts their scalability to more realistic settings where train and test data are drawn from different distributions. To address these limitations, we present a novel DIL approach based on a mixture of prompt-tuned CLIP models (MoP-CLIP), which generalizes the paradigm of S-Prompting to handle both in-distribution and out-of-distribution data at inference. In particular, at the training stage we model the features distribution of every class in each domain, learning individual text and visual prompts to adapt to a given domain. At inference, the learned distributions allow us to identify whether a given test sample belongs to a known domain, selecting the correct prompt for the classification task, or from an unseen domain, leveraging a mixture of the prompt-tuned CLIP models. Our empirical evaluation reveals the poor performance of existing DIL methods under domain shift, and suggests that the proposed MoP-CLIP performs competitively in the standard DIL settings while outperforming state-of-the-art methods in OOD scenarios. These results demonstrate the superiority of MoP-CLIP, offering a robust and general solution to the problem of domain incremental learning." @default.
- W4384263460 created "2023-07-14" @default.
- W4384263460 creator A5001640643 @default.
- W4384263460 creator A5004770604 @default.
- W4384263460 creator A5017608284 @default.
- W4384263460 creator A5051757079 @default.
- W4384263460 creator A5060775497 @default.
- W4384263460 creator A5081963363 @default.
- W4384263460 date "2023-07-11" @default.
- W4384263460 modified "2023-10-17" @default.
- W4384263460 title "MoP-CLIP: A Mixture of Prompt-Tuned CLIP Models for Domain Incremental Learning" @default.
- W4384263460 doi "https://doi.org/10.48550/arxiv.2307.05707" @default.
- W4384263460 hasPublicationYear "2023" @default.
- W4384263460 type Work @default.
- W4384263460 citedByCount "0" @default.
- W4384263460 crossrefType "posted-content" @default.
- W4384263460 hasAuthorship W4384263460A5001640643 @default.
- W4384263460 hasAuthorship W4384263460A5004770604 @default.
- W4384263460 hasAuthorship W4384263460A5017608284 @default.
- W4384263460 hasAuthorship W4384263460A5051757079 @default.
- W4384263460 hasAuthorship W4384263460A5060775497 @default.
- W4384263460 hasAuthorship W4384263460A5081963363 @default.
- W4384263460 hasBestOaLocation W43842634601 @default.
- W4384263460 hasConcept C105795698 @default.
- W4384263460 hasConcept C119857082 @default.
- W4384263460 hasConcept C127413603 @default.
- W4384263460 hasConcept C134306372 @default.
- W4384263460 hasConcept C138885662 @default.
- W4384263460 hasConcept C154945302 @default.
- W4384263460 hasConcept C185592680 @default.
- W4384263460 hasConcept C198531522 @default.
- W4384263460 hasConcept C201995342 @default.
- W4384263460 hasConcept C27158222 @default.
- W4384263460 hasConcept C2776214188 @default.
- W4384263460 hasConcept C2777212361 @default.
- W4384263460 hasConcept C2780451532 @default.
- W4384263460 hasConcept C33923547 @default.
- W4384263460 hasConcept C36503486 @default.
- W4384263460 hasConcept C41008148 @default.
- W4384263460 hasConcept C41895202 @default.
- W4384263460 hasConcept C43617362 @default.
- W4384263460 hasConcept C48044578 @default.
- W4384263460 hasConcept C7149132 @default.
- W4384263460 hasConcept C77088390 @default.
- W4384263460 hasConceptScore W4384263460C105795698 @default.
- W4384263460 hasConceptScore W4384263460C119857082 @default.
- W4384263460 hasConceptScore W4384263460C127413603 @default.
- W4384263460 hasConceptScore W4384263460C134306372 @default.
- W4384263460 hasConceptScore W4384263460C138885662 @default.
- W4384263460 hasConceptScore W4384263460C154945302 @default.
- W4384263460 hasConceptScore W4384263460C185592680 @default.
- W4384263460 hasConceptScore W4384263460C198531522 @default.
- W4384263460 hasConceptScore W4384263460C201995342 @default.
- W4384263460 hasConceptScore W4384263460C27158222 @default.
- W4384263460 hasConceptScore W4384263460C2776214188 @default.
- W4384263460 hasConceptScore W4384263460C2777212361 @default.
- W4384263460 hasConceptScore W4384263460C2780451532 @default.
- W4384263460 hasConceptScore W4384263460C33923547 @default.
- W4384263460 hasConceptScore W4384263460C36503486 @default.
- W4384263460 hasConceptScore W4384263460C41008148 @default.
- W4384263460 hasConceptScore W4384263460C41895202 @default.
- W4384263460 hasConceptScore W4384263460C43617362 @default.
- W4384263460 hasConceptScore W4384263460C48044578 @default.
- W4384263460 hasConceptScore W4384263460C7149132 @default.
- W4384263460 hasConceptScore W4384263460C77088390 @default.
- W4384263460 hasLocation W43842634601 @default.
- W4384263460 hasOpenAccess W4384263460 @default.
- W4384263460 hasPrimaryLocation W43842634601 @default.
- W4384263460 hasRelatedWork W2963058055 @default.
- W4384263460 hasRelatedWork W2993065420 @default.
- W4384263460 hasRelatedWork W3037880068 @default.
- W4384263460 hasRelatedWork W3136652852 @default.
- W4384263460 hasRelatedWork W3174101157 @default.
- W4384263460 hasRelatedWork W4200511449 @default.
- W4384263460 hasRelatedWork W4287234591 @default.
- W4384263460 hasRelatedWork W4287753411 @default.
- W4384263460 hasRelatedWork W4307964207 @default.
- W4384263460 hasRelatedWork W4375852175 @default.
- W4384263460 isParatext "false" @default.
- W4384263460 isRetracted "false" @default.
- W4384263460 workType "article" @default.