Verifiable Reasoning for LLM-based Generative Recommendation
This work improves the reliability and effectiveness of LLM-based generative recommendation systems for users by introducing a verification step to mitigate reasoning degradation.
This paper addresses reasoning degradation in LLM-based generative recommendation, where existing methods suffer from homogeneous or error-accumulated reasoning due to a lack of intermediate verification. The authors propose a novel "reason-verify-recommend" paradigm that interleaves reasoning with verification, leading to substantial enhancements in recommendation effectiveness and scalability across four real-world datasets.
Reasoning in Large Language Models (LLMs) has recently shown strong potential in enhancing generative recommendation through deep understanding of complex user preference. Existing approaches follow a {reason-then-recommend} paradigm, where LLMs perform step-by-step reasoning before item generation. However, this paradigm inevitably suffers from reasoning degradation (i.e., homogeneous or error-accumulated reasoning) due to the lack of intermediate verification, thus undermining the recommendation. To bridge this gap, we propose a novel \textbf{\textit{reason-verify-recommend}} paradigm, which interleaves reasoning with verification to provide reliable feedback, guiding the reasoning process toward more faithful user preference understanding. To enable effective verification, we establish two key principles for verifier design: 1) reliability ensures accurate evaluation of reasoning correctness and informative guidance generation; and 2) multi-dimensionality emphasizes comprehensive verification across multi-dimensional user preferences. Accordingly, we propose an effective implementation called VRec. It employs a mixture of verifiers to ensure multi-dimensionality, while leveraging a proxy prediction objective to pursue reliability. Experiments on four real-world datasets demonstrate that VRec substantially enhances recommendation effectiveness and scalability without compromising efficiency. The codes can be found at https://github.com/Linxyhaha/Verifiable-Rec.