Token-Controlled Re-ranking for Sequential Recommendation via LLMs

2025-11-21

AI Summary

Existing LLM-based re-ranking methods lack fine-grained user control, struggling to jointly accommodate inherent user preferences and multi-attribute constraints; hard filtering often narrows the candidate pool excessively and degrades recommendation quality, leaving users in a passive reception role. This paper proposes COREC, the first framework leveraging large language models for *controllable sequential re-ranking*. COREC introduces a token-augmentation mechanism enabling fine-grained, predictable attribute-level instruction control, and jointly models explicit attribute signals with implicit preference representations to dynamically balance constraint satisfaction and personalized quality. Evaluated on multiple benchmark datasets, COREC significantly improves standard ranking metrics (e.g., NDCG@10) and substantially outperforms state-of-the-art methods in attribute compliance rate—thereby enabling active, collaborative recommendation.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Data Mining & Knowledge Management: Recommender Systems
• Reasoning under Uncertainty: Sequential Decision Making

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
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

The widespread adoption of Large Language Models (LLMs) as re-rankers is shifting recommender systems towards a user-centric paradigm. However, a significant gap remains: current re-rankers often lack mechanisms for fine-grained user control. They struggle to balance inherent user preferences with multiple attribute-based constraints, often resorting to simplistic hard filtering that can excessively narrow the recommendation pool and yield suboptimal results. This limitation leaves users as passive recipients rather than active collaborators in the recommendation process. To bridge this gap, we propose COREC, a novel token-augmented re-ranking framework that incorporates specific user requirements in co-creating the recommendation outcome. COREC empowers users to steer re-ranking results with precise and flexible control via explicit, attribute-based signals. The framework learns to balance these commands against latent preferences, yielding rankings that adhere to user instructions without sacrificing personalization. Experiments show that COREC: (1) exceeds state-of-the-art baselines on standard recommendation effectiveness and (2) demonstrates superior adherence to specific attribute requirements, proving that COREC enables fine-grained and predictable manipulation of the rankings.

Problem

• Balancing user preferences with attribute constraints in recommendations
• Enabling fine-grained user control over re-ranking processes
• Preventing excessive narrowing of recommendation pools through hard filtering

Innovation

• Token-augmented re-ranking framework for user control
• Balances explicit attribute commands with latent preferences
• Enables fine-grained predictable manipulation of rankings