Pıer
TidesCurrentsHarbor LightsLabBottlesAshore
Pıer

Navigation

  • Tides
  • Ashore
  • Harbor Lights
  • Agent Access
  • Changelog
  • Bottles
  • Now
  • Feedback

External links

GitHubCloudborne ↗

© 2026 Pier.

WatchingResearchWatching0 independent reports0

CanniUplift: A Holistic Framework for Mitigating Seller and Incentive Cannibalization in E-commerce Uplift Modeling

First seen · 7/6/2026, 11:52 PMLatest activity · 7/6/2026, 11:52 PM

CanniUplift addresses two forms of cannibalization that can distort e-commerce uplift modeling: incentives shifting spending across sellers without increasing platform GMV, and organic conversions or alternative rewards creating attribution noise. The framework combines Platform-level Global Alignment (PGA), Redemption-based Decomposition Denoising (RDD), and a Treat-Attention mechanism for treatment-user interactions. According to the abstract, experiments on synthetic and large-scale industrial datasets improved wAUUC and wQINI over state-of-the-art baselines. An online deployment reported a 4.08% relative increase in platform-wide incremental GMV over the production baseline, alongside improved ROI in online A/B tests.

Event heat · last 24 hours

No heat snapshots are available in the last 24 hours.

No heat snapshots are available in the last 24 hours.

Reporting Timeline

  1. AggregatorarXiv7/6, 11:52 PMnot independentRepresentative
    CanniUplift: A Holistic Framework for Mitigating Seller and Incentive Cannibalization in E-commerce Uplift Modeling