sdc17 / CrossGET

[ICML 2024] CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers.

https://arxiv.org/pdf/2305.17455.pdf

efficient-deep-learning model-acceleration multimodal-learning vision-language-transformer image-captioning visual-question-answering visual-reasoning framework transformer image-text-retrieval text-image-retrieval token-ensemble token-matching

CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers

Official implementation of CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers. Please stay tuned.

About

[ICML 2024] CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers.

https://arxiv.org/pdf/2305.17455.pdf

efficient-deep-learning model-acceleration multimodal-learning vision-language-transformer image-captioning visual-question-answering visual-reasoning framework transformer image-text-retrieval text-image-retrieval token-ensemble token-matching

BSD 3-Clause "New" or "Revised" License