Reproduce “Multimodal Sentiment Analysis Based on Causal Reasoning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Fuhai Chen, Pengpeng Huang, Xuri Ge, Jie Huang, Zishuo Bao
The authors present CF-MSA, a method for reducing modality bias in sentiment classification of image–text posts. It compares predictions from joint inputs with predictions when text or image is masked, then uses the contrasts to adjust scores. On two MVSA datasets, text-bias removal improves accuracy over a multimodal-biased baseline, but F1 gains vary and causal identification is not established.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.