ALM-MTA combines front-door causal identification with an adversarially learned mediator and contrastive learning to perform multi-touch attribution, reporting lifts of 0.04% DAU, 0.6% daily active creators, and 670% unit exposure efficiency on a 400M-DAU platform.
Causal meta-learning with multi-view graphs for cold-start recommendation.ACM Transactions on Knowledge Discovery from Data, 2025a
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ALM-MTA:Front-Door Causal Multi-Touch Attribution Method for Creator-Ecosystem Optimization
ALM-MTA combines front-door causal identification with an adversarially learned mediator and contrastive learning to perform multi-touch attribution, reporting lifts of 0.04% DAU, 0.6% daily active creators, and 670% unit exposure efficiency on a 400M-DAU platform.