SoftSkip: Empowering multi-modal dynamic pruning for single-stage referring comprehension

• Main Authors: WEERAKOON, Dulanga, SUBBARAJU, Vigneshwaran, TRAN, Tuan, MISRA, Archan
• 格式: text
• 語言: English
• 出版: Institutional Knowledge at Singapore Management University 2022
• 主題: Human-Robot Interaction; Referring Expression Comprehension; Pruning; Computer Vision; Natural Language Processing; Computer Engineering; Software Engineering
• 在線閱讀: https://ink.library.smu.edu.sg/sis_research/7707; https://ink.library.smu.edu.sg/context/sis_research/article/8710/viewcontent/multimedia_final.pdf

Supporting real-time referring expression comprehension (REC) on pervasive devices is an important capability for human-AI collaborative tasks. Model pruning techniques, applied to DNN models, can enable real-time execution even on resource-constrained devices. However, existing pruning strategies a...