Code and Data Repository for Two-Stage Dynamic Fusion Framework for Multimodal Classification Tasks
This repository provides a new image-text multimodal fusion framework designed for classification tasks. Four online datasets are provided to replicate the results in the paper. The results for MVSA and CrisisMMD datasets are located in the emc_data folder. For other datasets, please refer to the stage2/README.md for instructions.
Authors
- Ying Chen (ORCID: https://orcid.org/0000-0002-0366-131X)
- Shoumeng Ge
Publication Details
- Journal
- INFORMS journal on computing
- Published
- 2025-05-14
- DOI
- https://doi.org/10.1287/ijoc.2023.0448.cd
- Primary Topic
- Speech and dialogue systems
- Type
- article
- Field-Weighted Citation Impact
- 0.00