Saved in:
Bibliographic Details
Main Authors: Zhang, Xinyue, Ding, Yuanhao, Ao, Xiang
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.23197
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918467783884800
author Zhang, Xinyue
Ding, Yuanhao
Ao, Xiang
author_facet Zhang, Xinyue
Ding, Yuanhao
Ao, Xiang
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2604_23197
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Follow the TRACE: Exploiting Post-Click Trajectories for Online Delayed Conversion Rate Prediction
Zhang, Xinyue
Ding, Yuanhao
Ao, Xiang
Machine Learning
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.
title Follow the TRACE: Exploiting Post-Click Trajectories for Online Delayed Conversion Rate Prediction
topic Machine Learning
url https://arxiv.org/abs/2604.23197