LettinGo: Explore User Profile Generation for Recommendation System
Lu Wang, Di Zhang, Fangkai Yang, Pu Zhao, Jianfeng Liu, Yuefeng Zhan, Hao Sun, Qingwei Lin, Weiwei Deng, Dongmei Zhang, Feng Sun, Qi Zhang
Abstract
User profiling is pivotal for recommendation systems, as it transforms raw user interaction data into concise and structured representations that drive personalized recommendations. While traditional embedding-based profiles lack interpretability and adaptability, recent advances with large language models (LLMs) enable text-based profiles that are semantically richer and more transparent. However, existing methods often adhere to fixed formats that limit their ability to capture the full diversity of user behaviors. In this paper, we introduce LettinGo, a novel framework for generating diverse and adaptive user profiles. By leveraging the expressive power of LLMs and incorporating direct feedback from downstream recommendation tasks, our approach avoids the rigid constraints imposed by supervised fine-tuning (SFT). Instead, we employ Direct Preference Optimization (DPO) to align the profile generator with task-specific performance, ensuring that the profiles remain adaptive and effective. LettinGo operates in three stages: (1) exploring diverse user profiles via multiple LLMs(2) evaluating profile quality based on their impact in recommendation systems, and (3) aligning the profile generation through pairwise preference data derived from task performance. Experimental results demonstrate that our framework significantly enhances recommendation accuracy, flexibility, and contextual awareness. This work enhances profile generation as a key innovation for next-generation recommendation systems.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
6 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- MUSE: A Multimodal Conversational Recommendation Dataset with Scenario-Grounded User Profiles cites LoRA: Low-Rank Adaptation of Large Language Models
- Enriching Semantic Profiles into Knowledge Graph for Recommender Systems Using Large Language Models cites Representation Learning with Large Language Models for Recommendation
- IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues cites LoRA: Low-Rank Adaptation of Large Language Models
- ProEx: A Unified Framework Leveraging Large Language Model with Profile Extrapolation for Recommendation cites LoRA: Low-Rank Adaptation of Large Language Models
- ProEx: A Unified Framework Leveraging Large Language Model with Profile Extrapolation for Recommendation cites Representation Learning with Large Language Models for Recommendation
Cited by top-tier papers6
- CanonFedRec: A Canonical Geometric Framework for Personalized Federated RecommendationYunqi Mi, Zeyu Hao, Guoshuai Zhao, Yingjie Wu et al.KDD 2026 · 2 citations
- Enriching Semantic Profiles into Knowledge Graph for Recommender Systems Using Large Language ModelsSeokho Ahn, Sungbok Shin, Young-Duk SeoKDD 2026 · 1 citation
- ProEx: A Unified Framework Leveraging Large Language Model with Profile Extrapolation for RecommendationYi Zhang, Yiwen Zhang, Yu Wang, Tong Chen et al.KDD 2026 · 1 citation
- Task-Aware Automated User Profile Generation for Recommendation Simulation Using Large Language ModelsXinye Wanyan, Chenglong Ma, Danula Hettiachchi, Ziqi Xu et al.SIGIR 2026
- How Humans Naturally Refer to Targets: Understanding Multimodal Instruction Patterns in Human-Robot InteractionLesong Jia, Makayla Chang, Yu Liu, Na DuCHI 2026
Builds on5
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu et al.ICLR 2022 · 18,833 citations
- Direct Preference Optimization: Your Language Model is Secretly a Reward ModelRafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning et al.NeurIPS 2023 · 10,924 citations
- Representation Learning with Large Language Models for RecommendationXubin Ren, Wei Wei, Lianghao Xia, Lixin Su et al.WWW 2024 · 385 citations
- IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner MonologuesDiji Yang, Jinmeng Rao, Kezhen Chen, Xiaoyuan Guo et al.SIGIR 2024 · 45 citations
- MUSE: A Multimodal Conversational Recommendation Dataset with Scenario-Grounded User ProfilesZihan Wang, Xiaocui Yang, Yongkang Liu, Shi Feng et al.ACL Findings 2025 · 5 citations
Related papers
- ProMax: Exploring the Potential of LLM-derived Profiles with Distribution Shaping for Recommender SystemsYi Zhang, Yiwen Zhang, Kai Zheng, Tong Chen et al.SIGIR 2026
- Transparent and Scrutable Recommendations Using Natural Language User ProfilesJerome Ramos, Hossein A. Rahmani, Xi Wang, Xiao Fu et al.ACL 2024
- Guided Profile Generation Improves Personalization with Large Language ModelsJiarui ZhangEMNLP Findings 2024 · 6 citations
- Large Language Models for Intent-Driven Session RecommendationsZhu Sun, Hongyang Liu, Xinghua Qu, Kaidong Feng et al.SIGIR 2024 · 35 citations
- CoT4Rec: Revealing User Preferences Through Chain of Thought for Recommender SystemsWeiqi Yue, Yuyu Yin, Xin Zhang, Binbin Shi et al.AAAI 2025 · 10 citations
