ManCAR：基於流形約束潛在推理與自適應測試時計算的序列推薦方法

ManCAR: Manifold-Constrained Latent Reasoning with Adaptive Test-Time Computation for Sequential Recommendation

February 23, 2026

作者: Kun Yang, Yuxuan Zhu, Yazhe Chen, Siyao Zheng, Bangyang Hong, Kangle Wu, Yabo Ni, Anxiang Zeng, Cong Fu, Hui Li

cs.AI

摘要

順序推薦系統日益採用潛在多步推理來增強測試時計算效率。儘管實證效果顯著，現有方法大多通過目標主導的目標驅動中間推理狀態，卻未施加明確的可行性約束。這導致潛在偏移現象，即推理軌跡偏離至不合理區域。我們主張有效的推薦推理應被視為在協作流形上的導航，而非自由形式的潛在優化。為此，我們提出ManCAR（流形約束自適應推理），這是一個將推理過程錨定於全局交互圖拓撲結構的理論框架。ManCAR從用戶近期行為的協作鄰域構建局部意圖先驗，將其表示為項目單純形上的概率分佈。在訓練過程中，模型逐步將其潛在預測分佈與此先驗對齊，迫使推理軌跡始終保持在有效流形內。測試時，推理會自適應進行直至預測分佈穩定，避免過度優化。我們提供ManCAR的變分解釋，從理論上驗證其偏移預防機制和自適應測試終止機制。在七個基準數據集上的實驗表明，ManCAR持續優於現有頂尖基準模型，在NDCG@10指標上實現最高46.88%的相對提升。代碼已開源於：https://github.com/FuCongResearchSquad/ManCAR。

English

Sequential recommendation increasingly employs latent multi-step reasoning to enhance test-time computation. Despite empirical gains, existing approaches largely drive intermediate reasoning states via target-dominant objectives without imposing explicit feasibility constraints. This results in latent drift, where reasoning trajectories deviate into implausible regions. We argue that effective recommendation reasoning should instead be viewed as navigation on a collaborative manifold rather than free-form latent refinement. To this end, we propose ManCAR (Manifold-Constrained Adaptive Reasoning), a principled framework that grounds reasoning within the topology of a global interaction graph. ManCAR constructs a local intent prior from the collaborative neighborhood of a user's recent actions, represented as a distribution over the item simplex. During training, the model progressively aligns its latent predictive distribution with this prior, forcing the reasoning trajectory to remain within the valid manifold. At test time, reasoning proceeds adaptively until the predictive distribution stabilizes, avoiding over-refinement. We provide a variational interpretation of ManCAR to theoretically validate its drift-prevention and adaptive test-time stopping mechanisms. Experiments on seven benchmarks demonstrate that ManCAR consistently outperforms state-of-the-art baselines, achieving up to a 46.88% relative improvement w.r.t. NDCG@10. Our code is available at https://github.com/FuCongResearchSquad/ManCAR.