Research map / Label-efficient and robust learning / Recognition and long-tail
Capturing Delayed Feedback in Conversion Rate Prediction via Elapsed-Time Sampling
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use elapsed-time sampling and importance-weighted loss | Streaming CVR prediction · Criteo Dataset | AUC: 0.8376 -> 0.8402* | Helped |
| Use elapsed-time sampling and importance-weighted loss | Streaming CVR prediction · Criteo Dataset | PR-AUC: 0.6313 -> 0.6393* | Helped |
| Use elapsed-time sampling and importance-weighted loss | Streaming CVR prediction · Criteo Dataset | NLL: 0.4033 -> 0.3924* | Helped |
| Use elapsed-time sampling and importance-weighted loss | Streaming CVR prediction · Taobao Dataset | AUC: 0.8851 -> 0.8895* | Helped |
| Use elapsed-time sampling and importance-weighted loss | Streaming CVR prediction · Taobao Dataset | PR-AUC: 0.6672 -> 0.6762* | Helped |
| Use elapsed-time sampling and importance-weighted loss | Streaming CVR prediction · Taobao Dataset | NLL: 0.1137 -> 0.1112* | Helped |
| Tune elapsed time to around 15 minutes | ES-DFM · Criteo Dataset | Performance | Helped |
| Use elapsed time larger than 1 hour | ES-DFM · Criteo Dataset | Performance | Hurt |
| Increase disturbance strength by randomly swapping positive labels with negative ones | Streaming CVR prediction methods · Robustness experiment | NLL | Mixed |
| Use ES-DFM instead of the best baseline | Online CVR prediction · Online A/B test, 7 days | AUC: 0.3% | Helped |
| Use ES-DFM instead of the best baseline | Online CVR prediction · Online A/B test, 7 days | CVR: 0.7% | Helped |
| Use ES-DFM instead of the best baseline | Online CVR prediction · Online A/B test, 7 days | GMV: 1.8% | Helped |
Closest papers by meaning
- Online Conversion Rate Prediction via Multi-Interval Screening and Synthesizing under Delayed Feedback (AAAI 2024 · similarity 0.93)
- Entire-Space Variational Information Exploitation for Post-Click Conversion Rate Prediction (AAAI 2025 · similarity 0.84)
- Percentile Risk-Constrained Budget Pacing for Guaranteed Display Advertising in Online Optimization (AAAI 2024 · similarity 0.83)
- CCS: Controllable and Constrained Sampling with Diffusion Models via Initial Noise Perturbation (NeurIPS 2025 · similarity 0.82)
- Jump Your Steps: Optimizing Sampling Schedule of Discrete Diffusion Models (ICLR 2025 · similarity 0.82)
- A Simple Early Exiting Framework for Accelerated Sampling in Diffusion Models (ICML 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.