Yu-Han Li

Precision at Scale: An End-to-End Graph-based Framework for Mitigating Network Interference in TikTok A/B Tests

This paper presents a production-ready framework deployed at TikTok, which integrates three core contributions to address the challenges of large-scale A/B tests on social platforms, and reduces interference rates by 68.8%, correcting a biased treatment effect estimate and enables previously undetectable cross-ecosystem measurements.