analiza aiqbench
EPIC advances generative recommendation by integrating explicit item-level competition into the denoising process of Semantic ID diffusion models, addressing the ambiguity where partial token sequences map to multiple catalog items. By projecting a personalized posterior distribution derived from user history back onto unresolved positions, this method guides token decisions without freezing the pretrained backbone or requiring extra decoder passes. The approach demonstrates consistent performance gains on Amazon benchmarks, highlighting how refining the generative trajectory improves next-item prediction accuracy.
EPIC: Explicit Posterior Item Conditioning for Semantic ID Diffusion Recommendation
arXiv:2609.03522v1 Announce Type: cross Abstract: 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.
Abstract: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.
Information Retrieval (cs.IR); Machine Learning (cs.LG)
arXiv:2609.03522 [cs.IR]
(or arXiv:2609.03522v1 [cs.IR] for this version)
https://doi.org/10.48550/arXiv.2609.03522 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 248 cuvinte
Subiecte conexe
PTL-Diffusion: Manifold-Aware Diffusion with Periodic Terminal Laws
arXiv · 9 iun. 2026
Provable diffusion-based posterior sampling for linear inverse problems via DDIM
arXiv · 23 iul. 2026
Adaptive Item-based Collaborative Structures via Noise Rescheduling in Diffusion for Generative Recommendation
arXiv · 25 aug. 2026
DiffCold: A Diffusion-based Generative Model for Cold-Start Item Recommendation
arXiv · 11 iun. 2026