Token-Weighted Multi-Target Learning for Generative Recommenders with Curriculum Learning

TL;DR

This paper introduces a novel token-weighted multi-target learning framework with curriculum learning for generative recommender systems, improving performance by emphasizing informative tokens and balancing popularity bias.

Contribution

It proposes two new token-weighting strategies based on information gain and integrates them into a multi-target curriculum learning framework for generative recommenders.

Findings

Outperforms existing methods on benchmark datasets.

Enhances robustness and generalization across semantic-ID types.

Achieves significant improvements on both popular and rare items.

Abstract

Generative recommender systems have recently attracted attention by formulating next-item prediction as an autoregressive sequence generation task. However, most existing methods optimize standard next-token likelihood and implicitly treat all tokens as equally informative, which is misaligned with semantic-ID-based generation. Accordingly, we propose two complementary information-gain-based token-weighting strategies tailored to generative recommendation with semantic IDs. Front-Greater Weighting captures conditional semantic information gain by prioritizing early tokens that most effectively reduce candidate-item uncertainty given their prefixes and encode coarse semantics. Frequency Weighting models marginal information gain under long-tailed item and token distributions, upweighting rare tokens to counteract popularity bias. Beyond individual strategies, we introduce a multi-target learning framework with curriculum learning that jointly optimizes the two token-weighted objectives alongside standard likelihood, enabling stable optimization and adaptive emphasis across training stages. Extensive experiments on benchmark datasets show that our method consistently outperforms strong baselines and existing token-weighting approaches, with improved robustness, strong generalization across different semantic-ID constructions, and substantial gains on both head and tail items. Code is available at https://github.com/CHIUWEINING/Token-Weighted-Multi-Target-Learning-for-Generative-Recommenders-with-Curriculum-Learning.