STAMImputer uses a mixture-of-experts framework with low-rank guided graph attention to impute missing traffic data, reporting small MAE improvements over prior methods on four benchmarks.
Saits: Self-attention-based imputation for time series.Expert Systems with Applications, 219:119619,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
STAMImputer: Spatio-Temporal Attention MoE for Traffic Data Imputation
STAMImputer uses a mixture-of-experts framework with low-rank guided graph attention to impute missing traffic data, reporting small MAE improvements over prior methods on four benchmarks.