Search papers, labs, and topics across Lattice.
This paper introduces BARGE, a novel approach to Generative Recommendation that addresses structural gaps in traditional autoregressive models by employing Item Context-Aware Attention (ICA) for item-level encoding and Hierarchical Path Reranking (HPR) with Dual-Path Decoding (DPD) to mitigate semantic drift. The method significantly enhances recommendation performance, as evidenced by extensive experiments on public benchmarks and a large-scale offline test. Notably, an online A/B test on a Tencent platform showed substantial improvements in key engagement metrics, underscoring BARGE's practical effectiveness in real-world applications.
BARGE not only restores item-level structure in recommendations but also achieves a 1.70% increase in total reading time on a major platform, showcasing its industrial relevance.
Generative Recommendation (GR) has emerged as a new paradigm for sequential recommendation, in which a representative line of work encodes items into hierarchical semantic IDs via residual quantization and predicts the IDs token by token. However, this generative formulation still exhibits structural gaps with respect to the recommendation task: flattening multi-token IDs into a single sequence destroys item-level structure, and the inconsistency between training and inference over a hierarchical codebook gives rise to semantic drift. To bridge these two gaps, we propose BARGE, which employs Item Context-Aware Attention (ICA) to restore item-level structure during encoding, and Hierarchical Path Reranking (HPR) together with Dual-Path Decoding (DPD) to suppress semantic drift from two complementary angles during decoding. Extensive experiments and analytical studies on public benchmarks and a large-scale offline test demonstrate that BARGE achieves superior recommendation performance. An online A/B test on a Tencent platform yields improvements of 0.60% in click-through rate, 1.34% in click unique visitors, and 1.70% in total reading time, confirming the practical value of BARGE in industrial-scale recommendation.
