DTRec: Learning Dynamic Reasoning Trajectories for Sequential Recommendation

2025-12-15

AI Summary

Existing reasoning-enhanced sequential recommendation methods suffer from two critical limitations: static reasoning directionality—contradicting human hierarchical cognition—and fixed reasoning depth—ignoring user behavioral heterogeneity—resulting in suboptimal performance and computational redundancy. To address these, we propose a dynamic reasoning modeling framework. First, we introduce Hierarchical Process Supervision (HPS), the first method to guide interpretable, multi-step reasoning paths via coarse-to-fine hierarchical supervision signals. Second, we design an Adaptive Reasoning Halting (ARH) mechanism that dynamically determines termination based on three criteria: confidence, consistency, and convergence. Our approach integrates differentiable halting conditions, a dynamic step-size architecture, and an LLM-inspired multi-step reasoning paradigm. Extensive experiments on three real-world datasets demonstrate up to 24.5% improvement in recommendation accuracy and up to 41.6% reduction in computational overhead.

Technology Category

• Reasoning under Uncertainty: Sequential Decision Making
• Search and Optimization: Metareasoning and Metaheuristics
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

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: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Inspired by advances in LLMs, reasoning-enhanced sequential recommendation performs multi-step deliberation before making final predictions, unlocking greater potential for capturing user preferences. However, current methods are constrained by static reasoning trajectories that are ill-suited for the diverse complexity of user behaviors. They suffer from two key limitations: (1) a static reasoning direction, which uses flat supervision signals misaligned with human-like hierarchical reasoning, and (2) a fixed reasoning depth, which inefficiently applies the same computational effort to all users, regardless of pattern complexity. These rigidity lead to suboptimal performance and significant computational waste. To overcome these challenges, we propose DTRec, a novel and effective framework that explores the Dynamic reasoning Trajectory for Sequential Recommendation along both direction and depth. To guide the direction, we develop Hierarchical Process Supervision (HPS), which provides coarse-to-fine supervisory signals to emulate the natural, progressive refinement of human cognitive processes. To optimize the depth, we introduce the Adaptive Reasoning Halting (ARH) mechanism that dynamically adjusts the number of reasoning steps by jointly monitoring three indicators. Extensive experiments on three real-world datasets demonstrate the superiority of our approach, achieving up to a 24.5% performance improvement over strong baselines while simultaneously reducing computational cost by up to 41.6%.

Problem

• Dynamic reasoning trajectories for sequential recommendation
• Hierarchical supervision to align with human cognitive processes
• Adaptive depth adjustment to optimize computational efficiency

Innovation

• Dynamic reasoning trajectories for sequential recommendation
• Hierarchical Process Supervision for coarse-to-fine direction guidance
• Adaptive Reasoning Halting to adjust steps based on indicators