# Factorized Latent Reasoning for LLM-based Recommendation

arXiv: https://arxiv.org/abs/2604.26760 · HTML: https://mlsanity.com/paper/2604.26760/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in LLM reasoning for recommendation, one of 17 directions in LLM reasoning.
- Builds on 18 of its 44 references that are in the map.
- First posted on arXiv 29 Apr 2026.

## Placed in

- LLM reasoning / [LLM reasoning for recommendation](https://mlsanity.com/lr/llm-reasoning-for-recommendation/)

## Builds on (most cited first, 18 of 18)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024)
- [Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach](https://mlsanity.com/paper/2502.05171/) (2025)
- [Implicit Chain of Thought Reasoning via Knowledge Distillation](https://mlsanity.com/paper/2311.01460/) (2023)
- [Generative Recommendation: Towards Next-generation Recommender Paradigm](https://mlsanity.com/paper/2304.03516/) (2023)
- [Think Before Recommend: Unleashing the Latent Reasoning Power for Sequential Recommendation](https://mlsanity.com/paper/2503.22675/) (2025)
- [OneRec-Think: In-Text Reasoning for Generative Recommendation](https://mlsanity.com/paper/2510.11639/) (2025)
- [Reinforced Latent Reasoning for LLM-based Recommendation](https://mlsanity.com/paper/2505.19092/) (2025)
- [Towards Agentic Recommender Systems in the Era of Multimodal Large Language Models](https://mlsanity.com/paper/2503.16734/) (2025)
- [Reason4Rec: Deliberative User Preference Alignment of Large Language Models for Recommendation](https://mlsanity.com/paper/2502.02061/) (2025)
- [Reason-to-Recommend: Using Interaction-of-Thought Reasoning to Enhance LLM Recommendation](https://mlsanity.com/paper/2506.05069/) (2025)
- [A Survey of Foundation Model-Powered Recommender Systems: From Feature-Based, Generative to Agentic Paradigms](https://mlsanity.com/paper/2504.16420/) (2025)
- [Listwise Preference Diffusion Optimization for User Behavior Trajectories Prediction](https://mlsanity.com/paper/2511.00530/) (2025)
- [Think before Recommendation: Autonomous Reasoning-enhanced Recommender](https://mlsanity.com/paper/2510.23077/) (2025)
- [Semi-Online Knowledge Distillation](https://mlsanity.com/paper/2111.11747/) (2021)
- [Generative Chain of Behavior for User Trajectory Prediction](https://mlsanity.com/paper/2601.18213/) (2026)
- [Beyond Negative Transfer: Disentangled Preference-Guided Diffusion for Cross-Domain Sequential Recommendation](https://mlsanity.com/paper/2509.00389/) (2025)

## Built on by (most cited first, 3 of 3)

- [Seeing and Reflecting: Multimodal Memory-Enhanced Agent Collaboration for Recommendation](https://mlsanity.com/paper/2607.07108/) (2026)
- [Hierarchical Latent Reasoning for LLM-based Recommendation](https://mlsanity.com/paper/2607.27760/) (2026)
- [AUSO: Action-Level Unified Skill Optimization from Internalization to Utilization](https://mlsanity.com/paper/2608.21292/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have recently been adopted for recommendation by framing user preference modeling as a language generation problem. However, existing latent reasoning approaches typically represent user intent with a single latent vector, which struggles to capture the inherently multi-faceted nature of user preferences. We propose Factorized Latent Reasoning (FLR), a novel framework for LLM-based sequential recommendation that decomposes latent reasoning into multiple disentangled preference factors. FLR introduces a lightweight multi-factor attention module that iteratively refines a latent thought representation, where each factor attends to distinct aspects of the user's interaction history. To encourage diversity and specialization, we design orthogonality, attention diversity, and sparsity regularization objectives, and dynamically aggregate factor contributions for the final prediction. We further integrate FLR with an efficient reinforcement learning strategy based on group-relative policy optimization, enabling stable alignment directly in the latent reasoning space. Experiments on multiple benchmarks show that FLR consistently outperforms strong baselines while improving robustness and interpretability.