Follow the TRACE: Exploiting Post-Click Trajectories for Online Delayed Conversion Rate Prediction

arXiv cs.LG / 4/28/2026

📰 NewsSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper addresses a core challenge in online conversion rate (CVR) prediction: delayed feedback creates a trade-off between label accuracy and using fresh data.
It proposes TRACE, which models the evolution of post-click behavior over the observation window as a “feedback trajectory” and updates conversion posteriors by comparing accumulated feedback status rather than waiting for final outcomes.
TRACE avoids applying hard labels to unrevealed samples, enabling dynamic refinement of predictions as new feedback arrives.
To mitigate early-stage sparsity in trajectory observations, the authors introduce a reliability-gated retrospective completer that uses full-lifecycle data to guide posterior updates for unrevealed samples.
Experiments on real settings show TRACE outperforms state-of-the-art baselines, and the retrospective completion component improves existing (model-agnostic) systems.

Abstract

Delayed feedback poses a core challenge for online CVR prediction, forcing a trade-off between label accuracy and data freshness. Existing methods address this through delay modeling or sample reweighting, yet neglect how post-click behaviors evolve over the observation period. To overcome this limitation, we formalize this evolution as feedback trajectory and propose TRACE. Instead of forcing hard labels on unrevealed samples, our method evaluates how well the accumulated feedback status aligns with conversion versus non-conversion, dynamically refining posteriors without waiting for final outcomes. To counteract early-stage trajectory sparsity, we further design a reliability-gated retrospective completer that leverages full-lifecycle data to provide adaptive posterior guidance for unrevealed samples. Extensive experiments validate TRACE's superiority over state-of-the-art baselines and confirm the retrospective completion module as a model-agnostic enhancer for existing systems. Our code is available at https://github.com/LunaZhangxy/TRACE.