# Tagging the Thought: Unlocking Personalization Reasoning via Reinforcement Learning

arXiv: https://arxiv.org/abs/2509.23140 · HTML: https://mlsanity.com/paper/2509.23140/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 16 of its 21 references that are in the map.
- First posted on arXiv 27 Sep 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 16 of 16)

- [Unsupervised Dense Information Retrieval with Contrastive Learning](https://mlsanity.com/paper/2112.09118/) (2021)
- [LaMP: When Large Language Models Meet Personalization](https://mlsanity.com/paper/2304.11406/) (2023)
- [Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging](https://mlsanity.com/paper/2310.11564/) (2023)
- [Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy](https://mlsanity.com/paper/2507.01352/) (2025)
- [Optimization Methods for Personalizing Large Language Models through Retrieval Augmentation](https://mlsanity.com/paper/2404.05970/) (2024)
- [Pearl: Personalizing Large Language Model Writing Assistants with Generation-Calibrated Retrievers](https://mlsanity.com/paper/2311.09180/) (2023)
- [Personalized Pieces: Efficient Personalized Large Language Models through Collaborative Efforts](https://mlsanity.com/paper/2406.10471/) (2024)
- [HYDRA: Model Factorization Framework for Black-Box LLM Personalization](https://mlsanity.com/paper/2406.02888/) (2024)
- [Teach LLMs to Personalize -- An Approach inspired by Writing Education](https://mlsanity.com/paper/2308.07968/) (2023)
- [Measuring What Makes You Unique: Difference-Aware User Modeling for Enhancing LLM Personalization](https://mlsanity.com/paper/2503.02450/) (2025)
- [LLMs + Persona-Plug = Personalized LLMs](https://mlsanity.com/paper/2409.11901/) (2024)
- [Reasoning-Enhanced Self-Training for Long-Form Personalized Text Generation](https://mlsanity.com/paper/2501.04167/) (2025)
- [PrLM: Learning Explicit Reasoning for Personalized RAG via Contrastive Reward Optimization](https://mlsanity.com/paper/2508.07342/) (2025)
- [Beyond Static Testbeds: An Interaction-Centric Agent Simulation Platform for Dynamic Recommender Systems](https://mlsanity.com/paper/2505.16429/) (2025)
- [Extended Inductive Reasoning for Personalized Preference Inference from Behavioral Signals](https://mlsanity.com/paper/2505.18071/) (2025)
- [Reasoning Meets Personalization: Unleashing the Potential of Large Reasoning Model for Personalized Generation](https://mlsanity.com/paper/2505.17571/) (2025)

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

- [Think-While-Generating: On-the-Fly Reasoning for Personalized Long-Form Generation](https://mlsanity.com/paper/2512.06690/) (2025)
- [ViPER: Empowering the Self-Evolution of Visual Perception Abilities in Vision-Language Model](https://mlsanity.com/paper/2510.24285/) (2025)
- [Preference-Aware Rubric Learning for Personalized Evaluation](https://mlsanity.com/paper/2605.31545/) (2026)

## Abstract (arXiv)

Recent advancements have endowed Large Language Models with impressive general reasoning capabilities. However, these reasoning models often perform worse than non-reasoning models on personalization tasks. While some methods use outcome-based RL to improve personalization reasoning, they fail to supervise the reasoning process. As a result, models may reach correct answers through flawed reasoning chains, limiting further improvement. To address this, we propose TagPR, a novel framework that adds semantic tags to the reasoning process for step-by-step guidance. TagPR first automatically generates a structured, tagged dataset for Supervised Fine-Tuning. It then employs a multi-stage RL process guided by a composite reward signal, which integrates tag-based process supervision with a novel Personalization Reward Model with User Embeddings to achieve fine-grained alignment with user-specific logic. Extensive experiments on public LaMP, LongLaMP, PGraphRAG, and a self-constructed dataset demonstrate that our approach achieves state-of-the-art results, delivering an average improvement of 32.65% over the base model across all LaMP benchmark tasks. Our work demonstrates that tag-guided process supervision is an effective approach for personalization reasoning.