Method Drift›Parameter-efficient fine-tuning (LoRA family)
MC Dropout
Parameter-efficient fine-tuning (LoRA family)
superseded — cited as a baseline and beaten by newer methods
0 papers critique it · 2 beat it on benchmarks
Beaten on benchmarks
Head-to-head results where a newer method reports beating MC Dropout. Values are copied from the source paper's tables — verify against the cited paper.
- Variational Low-Rank Adaptation Using IVON
IVON@mean beats MC Dropout · Accuracy (ACC) [All datasets]
78.9 vs 76.3
- Variational Low-Rank Adaptation Using IVON
IVON@mean beats MC Dropout · Expected Calibration Error (ECE) [All datasets]
17.2 vs 20.2
- Variational Low-Rank Adaptation Using IVON
IVON beats MC Dropout · Expected Calibration Error (ECE) [All datasets]
10.6 vs 20.2
- LoRA-Ensemble: Efficient Uncertainty Modelling for Self-Attention Networks
LoRA-Ensemble beats MC Dropout · Accuracy [16-member ensemble on CIFAR-100]
82.5 vs 77.1
- LoRA-Ensemble: Efficient Uncertainty Modelling for Self-Attention Networks
LoRA-Ensemble beats MC Dropout · Accuracy [16-member ensemble on HAM10000]
88.0 vs 83.7