RCLRec: Reverse Curriculum Learning for Modeling Sparse Conversions in Generative Recommendation

TL;DR

RCLRec introduces a reverse curriculum learning framework for generative recommendation, enhancing sparse conversion modeling by constructing targeted subsequences to improve conversion supervision and recommendation performance.

Contribution

It proposes a novel reverse curriculum learning approach that provides additional supervision for conversions, addressing sparsity in generative recommendation models.

Findings

Achieves +2.09% advertising revenue in online tests.

Achieves +1.86% increase in orders in online deployment.

Outperforms baseline models on offline datasets.

Abstract

Conversion objectives in large-scale recommender systems are sparse, making them difficult to optimize. Generative recommendation (GR) partially alleviates data sparsity by organizing multi-type behaviors into a unified token sequence with shared representations, but conversion signals remain insufficiently modeled. While recent behavior-aware GR models encode behavior types and employ behavior-aware attention to highlight decision-related intermediate behaviors, they still rely on standard attention over the full history and provide no additional supervision for conversions, leaving conversion sparsity largely unresolved. To address these challenges, we propose RCLRec, a reverse curriculum learning-based GR framework for sparse conversion supervision. For each conversion target, RCLRec constructs a short curriculum by selecting a subsequence of conversion-related items from the history in reverse. Their semantic tokens are fed to the decoder as a prefix, together with the target conversion tokens, under a joint generation objective. This design provides additional instance-specific intermediate supervision, alleviating conversion sparsity and focusing the model on the user's critical decision process. We further introduce a curriculum quality-aware loss to ensure that the selected curricula are informative for conversion prediction. Experiments on offline datasets and an online A/B test show that RCLRec achieves superior performance, with +2.09% advertising revenue and +1.86% orders in online deployment.