A graph-attention transformer over a fixed accounting-relation graph forecasts 13 small-business financial KPIs more accurately than LightGBM, TimeMixer, SOFTS, and time-series foundation models on a company-disjoint panel.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Accounting Graph Transformer for Short-History Multi-KPI Forecasting in Small Businesses
A graph-attention transformer over a fixed accounting-relation graph forecasts 13 small-business financial KPIs more accurately than LightGBM, TimeMixer, SOFTS, and time-series foundation models on a company-disjoint panel.