Energy & materialsResearch findings
AIE: Auction Information Enhanced Framework for CTR Prediction
Original published 2024.08.15 12:25 KSTAdded to Blackdeck 2026.10.26 01:31 KST
What has changed?
First reveals auction bias and effectively utilizes posterior auction signals.
Key points
- Highlights the importance of auction information in CTR prediction.
- Effectively leverages posterior auction signals via AM2 and BCM modules.
- Achieves 5.76% and 2.44% improvements in eCPM and CTR in real-world A/B tests.
Connect it to your work
Improved CTR prediction by integrating auction information with bias awareness
What remains unproven
This research note is based on an abstract; check the original for test conditions and application scope.
Originals used for this note
Links open the original publisher in a new tab.