VALOR improves uplift modeling for B2B sales revenue by using a treatment-gated sparse network and cost-sensitive focal loss, achieving 20% better rankability on benchmarks and 2.7x incremental revenue in a production A/B test.
In: International conference on predictive applications and APIs
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
VALOR: Value-Aware Revenue Uplift Modeling with Treatment-Gated Representation for B2B Sales
VALOR improves uplift modeling for B2B sales revenue by using a treatment-gated sparse network and cost-sensitive focal loss, achieving 20% better rankability on benchmarks and 2.7x incremental revenue in a production A/B test.