MAC: A Conversion Rate Prediction Benchmark Featuring Labels Under Multiple Attribution Mechanisms

Dimensions overall score 9.0

GitHub Code Pulse

No public code linked for this paper yet.

• The Multi-Attribution Benchmark (MAC) is the first public CVR dataset featuring labels from multiple attribution mechanisms.technical 100%
• PyMAL is an open-source library developed to promote reproducible research on MAL, covering a wide array of baseline methods.technical 100%
• MAL brings consistent performance gains across different attribution settings, especially for users featuring long conversion paths.result 90%
• Adding auxiliary objectives can be counterproductive when predicting first-click conversion targets, requiring careful selection of objectives.result 90%
• Two architectural design principles are paramount: fully learn the multi-attribution knowledge, and fully leverage this knowledge to serve the main task.technical 90%
• Mixture of Asymmetric Experts (MoAE) substantially surpasses the existing state-of-the-art MAL method on the MAC benchmark.result 95%
• The accuracy of attribution models like DDA may vary, potentially limiting the benchmark's effectiveness.limitation 70%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card