Open Access

ARTICLE

Enhancing Deep Learning Semantics: The Diffusion Sampling and Label-Driven Co-Attention Approach

Chunhua Wang^1,2, Wenqian Shang^1,2,*, Tong Yi^3,*, Haibin Zhu⁴

1 State Key Laboratory of Media Convergence and Communication, Communication University of China, Beijing, 100024, China
2 School of Computer and Cyber Sciences, Communication University of China, Beijing, 100024, China
3 School of Computer Science and Engineering, Guangxi Normal University, Guilin, 541004, China
4 Department of Computer Science, Nipissing University, North Bay, ON P1B 8L7, Canada

* Corresponding Authors: Wenqian Shang. Email: ; Tong Yi. Email:

(This article belongs to the Special Issue: The Next-generation Deep Learning Approaches to Emerging Real-world Applications)

Computers, Materials & Continua 2024, 79(2), 1939-1956. https://doi.org/10.32604/cmc.2024.048135

Received 28 November 2023; Accepted 12 March 2024; Issue published 15 May 2024

Abstract

The advent of self-attention mechanisms within Transformer models has significantly propelled the advancement of deep learning algorithms, yielding outstanding achievements across diverse domains. Nonetheless, self-attention mechanisms falter when applied to datasets with intricate semantic content and extensive dependency structures. In response, this paper introduces a Diffusion Sampling and Label-Driven Co-attention Neural Network (DSLD), which adopts a diffusion sampling method to capture more comprehensive semantic information of the data. Additionally, the model leverages the joint correlation information of labels and data to introduce the computation of text representation, correcting semantic representation biases in the data, and increasing the accuracy of semantic representation. Ultimately, the model computes the corresponding classification results by synthesizing these rich data semantic representations. Experiments on seven benchmark datasets show that our proposed model achieves competitive results compared to state-of-the-art methods.

---

Keywords

---