# Lusifer: LLM-based User SImulated Feedback Environment for online Recommender systems

arXiv: https://arxiv.org/abs/2405.13362 · HTML: https://mlsanity.com/paper/2405.13362/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 28 references that are in the map.
- First posted on arXiv 22 May 2024.

## Builds on (most cited first, 15 of 15)

- [Neural Collaborative Filtering](https://mlsanity.com/paper/1708.05031/) (2017)
- [Session-based Recommendations with Recurrent Neural Networks](https://mlsanity.com/paper/1511.06939/) (2015)
- [A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications](https://mlsanity.com/paper/2402.07927/) (2024)
- [Character-LLM: A Trainable Agent for Role-Playing](https://mlsanity.com/paper/2310.10158/) (2023)
- [Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches](https://mlsanity.com/paper/1907.06902/) (2019)
- [Quantifying the Persona Effect in LLM Simulations](https://mlsanity.com/paper/2402.10811/) (2024)
- [User Behavior Simulation with Large Language Model based Agents](https://mlsanity.com/paper/2306.02552/) (2023)
- [Large Language Models are Superpositions of All Characters: Attaining Arbitrary Role-play via Self-Alignment](https://mlsanity.com/paper/2401.12474/) (2024)
- [MetaAgents: Large Language Model Based Agents for Decision-Making on Teaming](https://mlsanity.com/paper/2310.06500/) (2023)
- [Virtual-Taobao: Virtualizing Real-world Online Retail Environment for Reinforcement Learning](https://mlsanity.com/paper/1805.10000/) (2018)
- [RecoGym: A Reinforcement Learning Environment for the problem of Product Recommendation in Online Advertising](https://mlsanity.com/paper/1808.00720/) (2018)
- [Unlocking the Potential of User Feedback: Leveraging Large Language Model as User Simulator to Enhance Dialogue System](https://mlsanity.com/paper/2306.09821/) (2023)
- [USimAgent: Large Language Models for Simulating Search Users](https://mlsanity.com/paper/2403.09142/) (2024)
- [RecSim NG: Toward Principled Uncertainty Modeling for Recommender Ecosystems](https://mlsanity.com/paper/2103.08057/) (2021)
- [LLM-driven Imitation of Subrational Behavior : Illusion or Reality?](https://mlsanity.com/paper/2402.08755/) (2024)

## Built on by (most cited first, 5 of 5)

- [Towards Unifying Evaluation of Counterfactual Explanations: Leveraging Large Language Models for Human-Centric Assessments](https://mlsanity.com/paper/2410.21131/) (2024)
- [Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems](https://mlsanity.com/paper/2607.04433/) (2026)
- [PersonaLedger: Generating Realistic Financial Transactions with Persona Conditioned LLMs and Rule Grounded Feedback](https://mlsanity.com/paper/2601.03149/) (2026)
- [End-to-End Personalization: Unifying Recommender Systems with Large Language Models](https://mlsanity.com/paper/2508.01514/) (2025)
- [Vectorized Context-Aware Embeddings for GAT-Based Collaborative Filtering](https://mlsanity.com/paper/2510.26461/) (2025)

## Abstract (arXiv)

Reinforcement learning (RL) recommender systems often rely on static datasets that fail to capture the fluid, ever changing nature of user preferences in real-world scenarios. Meanwhile, generative AI techniques have emerged as powerful tools for creating synthetic data, including user profiles and behaviors. Recognizing this potential, we introduce Lusifer, an LLM-based simulation environment designed to generate dynamic, realistic user feedback for RL-based recommender training. In Lusifer, user profiles are incrementally updated at each interaction step, with Large Language Models (LLMs) providing transparent explanations of how and why preferences evolve. We focus on the MovieLens dataset, extracting only the last 40 interactions for each user, to emphasize recent behavior. By processing textual metadata (such as movie overviews and tags) Lusifer creates more context aware user states and simulates feedback on new items, including those with limited or no prior ratings. This approach reduces reliance on extensive historical data and facilitates cold start scenario handling and adaptation to out of distribution cases. Our experiments compare Lusifer with traditional collaborative filtering models, revealing that while Lusifer can be comparable in predictive accuracy, it excels at capturing dynamic user responses and yielding explainable results at every step. These qualities highlight its potential as a scalable, ethically sound alternative to live user experiments, supporting iterative and user-centric evaluations of RL-based recommender strategies. Looking ahead, we envision Lusifer serving as a foundational tool for exploring generative AI-driven user simulations, enabling more adaptive and personalized recommendation pipelines under real world constraints.