cs.AINov 16, 2026

Learning to Reason with Predictive Gains for Generative Recommendation

Authors: Zhuoyang Li, Teng Shi, Jiakai Tang, Weijie Yu, Guofu Xie, Xiao Zhang, Lei Kou, Yang Yu, 

Organizations: Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China · School of Artificial Intelligence and Data Science, University of International Business and Economics, Beijing, China · Alibaba International Digital Commercial Group, Hangzhou, China

Abstract

Reasoning-enhanced generative recommendation (GR) has attracted increasing attention for leveraging the reasoning capabilities of large language models (LLMs) to improve next-item prediction. However, existing methods primarily optimize reasoning through Group Relative Policy Optimization (GRPO) with rule-based rewards determined by final recommendation performance, which faces two key limitations: (i) unmeasured predictive contributions within reasoning, where the predictive contributions of individual reasoning steps are not directly reflected in optimization signals, and (ii) weakly discriminative learning signals within groups, where different reasoning trajectories within the group often receive indistinguishable rewards, limiting effective policy optimization. In this paper, we propose GainReason, a simple yet effective training framework that optimizes reasoning through Predictive Gain (PG). Specifically, PG measures changes in target Semantic ID (SID) log-likelihood induced by individual reasoning sentences, providing contribution-aware and discriminative rewards. We further theoretically establish the alignment between expected PG and the recommendation ranking objective. Building on this signal, we develop PG-based Reasoning Optimization (PGRO) to optimize reasoning generation with sentence-level rewards, supported by a custom causal attention mechanism for efficient PG computation. Additionally, an auxiliary PG-based Reasoning Utilization (PGRU) task adaptively strengthens reasoning-conditioned prediction to complement PGRO. Extensive experiments on three public benchmarks demonstrate that GainReason consistently outperforms competitive baselines and achieves substantial performance gains from reasoning.

Figures & tables

Explore similar work

CardsList
  1. Implicit Reasoning for Large Language Model-based Generative Recommendation

    Jun 12, 2026Yinhan He, Liam Collins, Bhuvesh Kumar +3Implicit Reasoning in Language ModelsLarge Language Model-Based Recommendation

  2. GR2: Generative Reasoning Re-ranker

    Feb 8, 2026Mingfu Liang, Yufei Li, Jay Xu +20RL for Language ModelsRL for Recommendation