IF-DFM uses influence functions to update a conversion-rate model for delayed and newly arrived labels, approaching retraining accuracy at a fraction of the compute cost.
In ICCV, 4869–4879
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
-
Delayed Feedback Modeling with Influence Functions
IF-DFM uses influence functions to update a conversion-rate model for delayed and newly arrived labels, approaching retraining accuracy at a fraction of the compute cost.