Climber-Pilot: A Non-Myopic Generative Recommendation Model Towards Better Instruction-Following

TL;DR

Climber-Pilot introduces a unified generative retrieval framework that enhances long-horizon prediction and instruction-following capabilities, significantly improving recommendation relevance and business metrics in large-scale industrial settings.

Contribution

The paper proposes TAMIP for mitigating myopia in generative retrieval and CGSA for integrating retrieval constraints directly into the model, advancing the state-of-the-art in instruction-following recommendation systems.

Findings

Achieved 4.24% improvement in core business metric at NetEase Cloud Music.

Demonstrated superior performance over existing baselines in offline and online evaluations.

Effectively incorporated business constraints into generative retrieval without additional inference steps.

Abstract

Generative retrieval has emerged as a promising paradigm in recommender systems, offering superior sequence modeling capabilities over traditional dual-tower architectures. However, in large-scale industrial scenarios, such models often suffer from inherent myopia: due to single-step inference and strict latency constraints, they tend to collapse diverse user intents into locally optimal predictions, failing to capture long-horizon and multi-item consumption patterns. Moreover, real-world retrieval systems must follow explicit retrieval instructions, such as category-level control and policy constraints. Incorporating such instruction-following behavior into generative retrieval remains challenging, as existing conditioning or post-hoc filtering approaches often compromise relevance or efficiency. In this work, we present Climber-Pilot, a unified generative retrieval framework to address both limitations. First, we introduce Time-Aware Multi-Item Prediction (TAMIP), a novel training paradigm designed to mitigate inherent myopia in generative retrieval. By distilling long-horizon, multi-item foresight into model parameters through time-aware masking, TAMIP alleviates locally optimal predictions while preserving efficient single-step inference. Second, to support flexible instruction-following retrieval, we propose Condition-Guided Sparse Attention (CGSA), which incorporates business constraints directly into the generative process via sparse attention, without introducing additional inference steps. Extensive offline experiments and online A/B testing at NetEase Cloud Music, one of the largest music streaming platforms, demonstrate that Climber-Pilot significantly outperforms state-of-the-art baselines, achieving a 4.24\% lift of the core business metric.