Generate What You Can Trust: Content Credibility in Generative Recommenders

2026-10-05

AI Summary

This study addresses the risk of users being exposed to misinformation due to generative recommendation systems overlooking content credibility. To this end, it proposes CreGR, the first credibility-aware generative recommendation model. CreGR establishes a generation framework that explicitly incorporates credibility by introducing credibility signal decoupling and an asymmetric masking strategy, jointly ensuring the trustworthiness and accuracy of recommended content during both the tokenization and generation stages. Technically, the model employs a credibility-aware tokenizer alongside a discrete diffusion-based generation mechanism. Extensive experiments conducted on three real-world datasets demonstrate that CreGR significantly enhances the credibility of recommendation systems, thereby validating the effectiveness of the proposed approach.

Abstract

Generative recommendation (GR) represents items with semantic IDs (i.e., discrete token sequences) and generates target item tokens as recommendations. Despite its promising results, existing methods predominantly optimize for accuracy while neglecting the credibility of the recommendations they generate. This oversight inevitably exposes users to uncredible content (e.g., fake news) with serious societal consequences, including user distrust, reputation harm to platforms, and broader social instability. To address this critical yet underexplored challenge, we propose CreGR, the first credible GR model that jointly tackles content credibility across the two core stages of GR: tokenization and generation. In the tokenization stage, we design a new credibility-aware tokenizer that explicitly encourages the model to learn discriminative tokens respectively for credible and uncredible items, thereby disentangling credibility signals at the token level. Building on this, in the generation stage, we propose a novel accuracy-preserving and credibility-oriented generator grounded in discrete diffusion. Specifically, we introduce an asymmetric masking probability reduction strategy that selectively diminishes the contribution of tokens associated with uncredible content to the generation process, while leaving tokens encoding user preference signals unaffected so as to preserve recommendation accuracy. Experiments on three real-world datasets demonstrate the effectiveness of CreGR.

Problem

• Generative Recommendation
• Content Credibility
• Semantic IDs
• Fake News

Innovation

• Generative Recommendation
• Content Credibility
• Credibility-aware Tokenizer
• Discrete Diffusion
• Asymmetric Masking