Search papers, labs, and topics across Lattice.
This paper introduces Explicit Posterior Item Conditioning (EPIC) to enhance Semantic ID generative recommendation by integrating item-level competition into the denoising process. By constructing a personalized posterior distribution over candidate items based on user interactions and generation context, EPIC effectively guides token predictions, leading to improved recommendation accuracy. Experiments across four Amazon benchmarks demonstrate that EPIC consistently outperforms strong baselines, primarily due to its ability to maintain promising item hypotheses during the denoising steps.
Personalizing item selection during denoising boosts recommendation accuracy by leveraging user interactions more effectively than traditional methods.
Semantic ID (SID) generative recommendation predicts the next item by generating a short tuple of discrete tokens. Recent masked-diffusion methods improve this process through bidirectional context and flexible decoding, yet recommendation ultimately requires selecting among complete catalog items. At each denoising step, a partial SID can correspond to multiple feasible items, while existing methods primarily reason through position-wise token predictions. We propose Explicit Posterior Item Conditioning (EPIC), which introduces explicit item-level competition into SID denoising. EPIC constructs a personalized posterior over feasible candidate items using the current generation context and the user's recent interactions, then projects this distribution back to unresolved SID positions to guide subsequent token decisions. The pretrained backbone remains frozen and requires no additional decoder forward pass. Experiments on four Amazon benchmarks show consistent improvements over strong baselines, while diagnostic analyses indicate that the gains primarily arise from personalized transition evidence that preserves promising item hypotheses during denoising.
