MoXGATE, a cross-attention model with learned modality weights, reports 95% accuracy on gastrointestinal adenocarcinoma subtype classification from TCGA multi-omic data, but the result lacks error bars and has experimental inconsistencies.
Islet g protein-coupled receptors as potential targets for treatment of type 2 diabetes
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
MoXGATE: Modality-aware cross-attention for multi-omic gastrointestinal cancer sub-type classification
MoXGATE, a cross-attention model with learned modality weights, reports 95% accuracy on gastrointestinal adenocarcinoma subtype classification from TCGA multi-omic data, but the result lacks error bars and has experimental inconsistencies.