Give Users the Wheel: Towards Promptable Recommendation Paradigm

2026-02-21

AI Summary

This work addresses the challenge that existing sequential recommendation models struggle to dynamically respond to users’ immediate intents via natural language prompts while preserving collaborative filtering signals. To this end, we propose DPR, a decoupled promptable sequential recommendation framework that uniquely integrates prompt-guided capabilities with collaborative signals in a disentangled manner. DPR achieves effective alignment between ID-based collaborative retrieval and semantic prompt spaces through a signal alignment fusion module, a Mixture-of-Experts (MoE) architecture with gradient disentanglement for positive and negative guidance, and a three-stage progressive alignment training strategy. Extensive experiments demonstrate that DPR significantly outperforms state-of-the-art methods on real-world datasets, maintaining competitive performance on standard sequential recommendation tasks while effectively supporting natural language prompt guidance.

Technology Category

• Natural Language Processing: Prompt Engineering / Prompting
• Reasoning under Uncertainty: Sequential Decision Making
• Machine Learning: Mixture of Experts (MoE)

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Conventional sequential recommendation models have achieved remarkable success in mining implicit behavioral patterns. However, these architectures remain structurally blind to explicit user intent: they struggle to adapt when a user's immediate goal (e.g., expressed via a natural language prompt) deviates from their historical habits. While Large Language Models (LLMs) offer the semantic reasoning to interpret such intent, existing integration paradigms force a dilemma: LLM-as-a-recommender paradigm sacrifices the efficiency and collaborative precision of ID-based retrieval, while Reranking methods are inherently bottlenecked by the recall capabilities of the underlying model. In this paper, we propose Decoupled Promptable Sequential Recommendation (DPR), a model-agnostic framework that empowers conventional sequential backbones to natively support Promptable Recommendation, the ability to dynamically steer the retrieval process using natural language without abandoning collaborative signals. DPR modulates the latent user representation directly within the retrieval space. To achieve this, we introduce a Fusion module to align the collaborative and semantic signals, a Mixture-of-Experts (MoE) architecture that disentangles the conflicting gradients from positive and negative steering, and a three-stage training strategy that progressively aligns the semantic space of prompts with the collaborative space. Extensive experiments on real-world datasets demonstrate that DPR significantly outperforms state-of-the-art baselines in prompt-guided tasks while maintaining competitive performance in standard sequential recommendation scenarios.

Problem

• Promptable Recommendation
• Sequential Recommendation
• User Intent
• Large Language Models
• Collaborative Filtering

Innovation

• Promptable Recommendation
• Decoupled Promptable Sequential Recommendation
• Mixture-of-Experts
• Semantic-Collaborative Alignment
• Model-Agnostic Framework