J. Li

Multimodal Fusion Comprehension Model for English Listening and Literature Appreciation Based on the Perceiver IO Architecture

Traditional English multimodal listening comprehension models suffer from semantic fragmentation and weak task transferability because they lack a unified cross-modal attention mechanism, have difficulty processing high-dimensional images and long-duration speech signals simultaneously, and often rely on loosely couple...