Posts by Collection

portfolio

Portfolio item number 2

Short description of portfolio item number 2

publications

AMCCL: Adaptive Multi-scale Convolution Fusion Network with Contrastive Learning for Multimodal Sentiment Analysis

This paper proposes AMCCL, a deep learning framework designed for Multimodal Sentiment Analysis (MSA) using adaptive multi-scale convolution and contrastive learning.

@inproceedings{yu2026amccl,
  title={AMCCL: Adaptive Multi-scale Convolution Fusion Network with Contrastive Learning for Multimodal Sentiment Analysis},
  author={Yu, Jiakang and Li, Mingxin and Deng, Hongtao and Gao, Wang and Zhu, Xun},
  booktitle={Pacific Rim International Conference on Artificial Intelligence},
  pages={505--512},
  year={2026}
}

CodeMNER: Vision-Language Models are Better Multimodal Named Entity Recognizers via Progressive Vision-Code Alignment

This paper proposes CodeMNER, a multimodal code generation framework for Multimodal Named Entity Recognition (MNER) via progressive vision-code alignment.

@inproceedings{yu2026codemner,
  title={CodeMNER: Vision-Language Models are Better Multimodal Named Entity Recognizers via Progressive Vision-Code Alignment},
  author={Yu, Jiakang and Huang, Shizhou and Chen, Xiaode and Deng, Hongtao and Gao, Wang and Zhu, Xun},
  booktitle={Proceedings of the 2026 International Conference on Multimedia Retrieval},
  pages={497–505},
  year={2026}
}

Jiakang Yu (余家康)